Our work provides an insight in volatility-based portfolio switching and calculating the returns in the event of rule based portfolio allocation. we have selected four indices namely:

• GSPC and FTSE for studying volatility-based portfolio switching in developed market condition and performance of indices based etfs(GSPC and FTSE can act as a proxy for etfs) in American and European market conditions.
• Similarly, HSI(China) will provide the knowledge about investors decision making pattern in a market which is transitioning from emerging to developed market.
• NSEI(India) has been selected to simulate an emerging market.
• We have also taken Gold and Gold prices to study their volatility movements and provide a more diversified solution for portfolio allocation.

Adding libraries

# Numerical Libraries
import numpy as np
import pandas as pd
import scipy
# Standard
import datetime, time, math, random
from itertools import permutations, combinations
from datetime import datetime, timedelta, date
import pandas_datareader.data as web
from matplotlib.dates import DateFormatter
import scipy.stats as stats
# Data Import Libraries
import yfinance as yf
import pandas_datareader as pdr
# Supress Warnings
import warnings
warnings.filterwarnings('ignore')
# Plots
import seaborn as sns
import networkx as nx
from pylab import plt, mpl
from pandas.plotting import scatter_matrix
from scipy.stats import kurtosis,norm,genpareto

EQUITY

• GSPC - S&P500(USA)- Developed market(We can assume investors will be more rational in their decision making)
• FTSE - UK - Developed market
• HSI- Hong Kong Stock Exchange,China - A market transitioning from emerging to a developed market.
• NSEI- Nifty50 India, An emerging market.

# Function to get closing price from Yahoo. Input tickers as in Yahoo URL and Start/End Date YYYY-MM-DD
def GetData(T, Start, End):
    '''
    T: List of Stock indices, as in Yahoo URL for Download e.g ^GSPC for S&P 500
    Start: Date in YYYY-MM-DD
    End: Date in YYYY-MM-DD
    '''
    Close_ser = []
    t1 = int(pd.to_datetime(Start).timestamp())
    t2 = int(pd.to_datetime(End).timestamp())
    for i in T:
        URL = 'https://query1.finance.yahoo.com/v7/finance/download/'+i + '?period1=' + str(int(t1)) + '&period2='+ str(int(t2)) + '&interval=1d&events=history&includeAdjustedClose=true'
        Fin_DF = pd.read_csv(URL, header = 0, encoding = 'utf-8')
        Fin_DT = pd.to_datetime(Fin_DF['Date'])
        Fin_DF.insert(1,'DateTime', Fin_DT, allow_duplicates = False)
        Fin_DF = Fin_DF.drop(['Date'], axis=1).set_index('DateTime')
        Close_ser.append(Fin_DF)
    return Close_ser

Function for getting closed price

def GetAdjCloseQuotes(T, Start, End):
    '''
    T: List of Stock indices, as in Yahoo URL for Download e.g ^GSPC for S&P 500
    Start: Date in YYYY-MM-DD
    End: Date in YYYY-MM-DD
    '''
    P = GetData(T, Start, End)
    Indices_DF = []
    for i in range(len(P)):
        Indices_DF.append(pd.DataFrame(P[i]['Adj Close'].values, index = P[i].index, columns = [T[i]]))
    return Indices_DF[0].join(Indices_DF[1:], how = 'inner')

Tickers = [ '^FTSE', '^GSPC', '^HSI','^NSEI','GLD','GOLD.AX',]
Start_Date = "2009-01-01"
End_Date = "2023-05-01"

multiData = GetAdjCloseQuotes(Tickers, Start_Date, End_Date)
multiData.head()

	^FTSE	^GSPC	^HSI	^NSEI	GLD	GOLD.AX
DateTime
2009-01-02	4561.799805	931.799988	15042.809570	3046.750000	86.230003	124.330002
2009-01-05	4579.600098	927.450012	15563.309570	3121.449951	84.480003	120.639999
2009-01-06	4638.899902	934.700012	15509.509766	3112.800049	85.129997	118.199997
2009-01-07	4507.500000	906.650024	14987.459961	2920.399902	82.750000	117.660004
2009-01-09	4448.500000	890.349976	14377.440430	2873.000000	83.919998	119.599998

# Calculating daily returns
return_multiData = multiData.pct_change().dropna()
return_multiData

	^FTSE	^GSPC	^HSI	^NSEI	GLD	GOLD.AX
DateTime
2009-01-05	0.003902	-0.004668	0.034601	0.024518	-0.020295	-0.029679
2009-01-06	0.012949	0.007817	-0.003457	-0.002771	0.007694	-0.020225
2009-01-07	-0.028326	-0.030010	-0.033660	-0.061809	-0.027957	-0.004568
2009-01-09	-0.013089	-0.017978	-0.040702	-0.016231	0.014139	0.016488
2009-01-12	-0.005013	-0.022564	-0.028269	-0.034772	-0.037655	0.011957
...	...	...	...	...	...	...
2023-04-21	0.001455	0.000903	-0.015749	-0.000023	-0.009888	0.002906
2023-04-24	-0.000240	0.000852	-0.005768	0.006772	0.003039	-0.004346
2023-04-26	-0.007533	-0.019591	-0.010154	0.003956	-0.000379	0.018188
2023-04-27	-0.002674	0.019566	0.004201	0.005695	0.000054	-0.001072
2023-04-28	0.004980	0.008253	0.002736	0.008370	0.000271	-0.005722

3216 rows × 6 columns

We have selected historical volatility for our study as it will easier for investment decision makers to comprehend as historical data is readily available.

# Daily Volatility
TRADING_DAYS = 252
ann_vol_portfolio = return_multiData.rolling(window=TRADING_DAYS).std()*np.sqrt(TRADING_DAYS)
ann_vol_portfolio = ann_vol_portfolio.dropna()
ann_vol_portfolio.head()

	^FTSE	^GSPC	^HSI	^NSEI	GLD	GOLD.AX
DateTime
2010-02-10	0.232625	0.277415	0.335965	0.348596	0.215220	0.232228
2010-02-11	0.232659	0.277507	0.334748	0.348095	0.214936	0.230634
2010-02-17	0.233344	0.278052	0.334885	0.348433	0.215652	0.230759
2010-02-18	0.231682	0.276385	0.333109	0.342569	0.213742	0.230807
2010-02-19	0.231327	0.275734	0.331550	0.342246	0.213427	0.230304

# Create a figure and axis
fig, ax1 = plt.subplots()
# Plot returns on the first y-axis (left)
ax1.plot(return_multiData['^FTSE'], color='b', label='Returns')
ax1.set_xlabel('Time')
ax1.set_ylabel('Returns', color='b')
ax1.tick_params('y', colors='b')

# Create a second y-axis (right) for plotting volatility
ax2 = ax1.twinx()
ax2.plot(ann_vol_portfolio['^FTSE'], color='r', label='Volatility')
ax2.set_ylabel('Volatility', color='r')
ax2.tick_params('y', colors='r')

# Add legends
lines1, labels1 = ax1.get_legend_handles_labels()
lines2, labels2 = ax2.get_legend_handles_labels()
lines = lines1 + lines2
labels = labels1 + labels2
ax1.legend(lines, labels, loc='upper left')

<matplotlib.legend.Legend at 0x7e1aecf57850>

# Sharpe_ratio
sharpe_ratio = return_multiData.mean()/ann_vol_portfolio
sharpe_ratio.tail()

	^FTSE	^GSPC	^HSI	^NSEI	GLD	GOLD.AX
DateTime
2023-04-21	0.001478	0.002264	0.000561	0.003981	0.001780	0.000194
2023-04-24	0.001499	0.002264	0.000561	0.003982	0.001780	0.000194
2023-04-26	0.001537	0.002258	0.000563	0.004000	0.001789	0.000193
2023-04-27	0.001537	0.002267	0.000567	0.004045	0.001796	0.000194
2023-04-28	0.001536	0.002267	0.000567	0.004047	0.001822	0.000195

# Cumulative Returns
cumprod_daily_return = (1 + return_multiData).cumprod()
cumprod_daily_return.head()

	^FTSE	^GSPC	^HSI	^NSEI	GLD	GOLD.AX
DateTime
2009-01-05	1.003902	0.995332	1.034601	1.024518	0.979705	0.970321
2009-01-06	1.016901	1.003112	1.031025	1.021679	0.987243	0.950696
2009-01-07	0.988097	0.973009	0.996321	0.958530	0.959643	0.946352
2009-01-09	0.975163	0.955516	0.955768	0.942972	0.973211	0.961956
2009-01-12	0.970275	0.933956	0.928749	0.910183	0.936565	0.973458

# heatmap of equity
sns.heatmap(multiData.corr(), cmap = 'coolwarm', annot = True)
plt.title('Correlations Heatmap')
plt.show()

fig, axes = plt.subplots(1, 3, sharex=True, figsize=(20,5))

sns.lineplot(data=ann_vol_portfolio, dashes=False, ax=axes[0])
axes[0].set_ylabel('Volatility')
axes[0].set_title('Portfolio Annualized Volatilty')


sns.lineplot(data=cumprod_daily_return, dashes=False, ax=axes[1])
axes[1].set_ylabel('Cumulative return')
axes[1].set_title('Indices Cumulative Returns')

sns.lineplot(data=sharpe_ratio, dashes=False, ax=axes[2])
axes[2].set_ylabel('sharpe_ratio')
axes[2].set_title('Indices sharpe_ratio')

Text(0.5, 1.0, 'Indices sharpe_ratio')

Testing volatility Scenarios

We have converted historical volatility for simpler comprehension and binning for impletmentation of rules.

ann_vol_portfolio= (ann_vol_portfolio * 100).round()
ann_vol_portfolio.head()

	^FTSE	^GSPC	^HSI	^NSEI	GLD	GOLD.AX
DateTime
2010-02-10	23.0	28.0	34.0	35.0	22.0	23.0
2010-02-11	23.0	28.0	33.0	35.0	21.0	23.0
2010-02-17	23.0	28.0	33.0	35.0	22.0	23.0
2010-02-18	23.0	28.0	33.0	34.0	21.0	23.0
2010-02-19	23.0	28.0	33.0	34.0	21.0	23.0

vol_bins = [0, 10, 20, 30, 40, 50, 60]
vol_labels = ['0-9','10-19','20-29','30-39','40-49','50-59']
ann_vol_portfolio_2=pd.DataFrame()
ann_vol_portfolio_2['ftse_group'] = pd.cut(ann_vol_portfolio['^FTSE'], bins=vol_bins, labels=vol_labels, right=False)
ann_vol_portfolio_2['gspc_group'] = pd.cut(ann_vol_portfolio['^GSPC'], bins=vol_bins, labels=vol_labels, right=False)
ann_vol_portfolio_2['hsi_group'] = pd.cut(ann_vol_portfolio['^HSI'], bins=vol_bins, labels=vol_labels, right=False)
ann_vol_portfolio_2['nsei_group'] = pd.cut(ann_vol_portfolio['^NSEI'], bins=vol_bins, labels=vol_labels, right=False)
ann_vol_portfolio_2['gld_group'] = pd.cut(ann_vol_portfolio['GLD'], bins=vol_bins, labels=vol_labels, right=False)
ann_vol_portfolio_2['gold.ax_group'] = pd.cut(ann_vol_portfolio['GOLD.AX'], bins=vol_bins, labels=vol_labels, right=False)
ann_vol_portfolio_2.tail()

	ftse_group	gspc_group	hsi_group	nsei_group	gld_group	gold.ax_group
DateTime
2023-04-21	10-19	20-29	30-39	10-19	10-19	10-19
2023-04-24	10-19	20-29	30-39	10-19	10-19	10-19
2023-04-26	10-19	20-29	30-39	10-19	10-19	10-19
2023-04-27	10-19	20-29	30-39	10-19	10-19	10-19
2023-04-28	10-19	20-29	30-39	10-19	10-19	10-19

# Histogram of equity
ann_vol_portfolio.plot.hist(bins=10);

We have taken five scenarios where we define different volatility levels based on variance of historical volatility dataframe and check for condition where succesive days volatility has breached previous days volatilty by a particular margin.

Since these values are arbitrary, investors are free to define their volatility levels for switching.

# Count
count_df_1 = ((ann_vol_portfolio < 15) & (ann_vol_portfolio.diff() <10)).sum()
count_df_2 = (((20 > ann_vol_portfolio) & (ann_vol_portfolio >= 15)) | ((20 > ann_vol_portfolio.diff()) & (ann_vol_portfolio.diff() > 15))).sum()
count_df_3 = (((25 > ann_vol_portfolio) & (ann_vol_portfolio >= 20)) | ((30 > ann_vol_portfolio.diff()) & (ann_vol_portfolio.diff() > 20))).sum()
count_df_4 = (((27 > ann_vol_portfolio) & (ann_vol_portfolio >= 25)) | ((15 > ann_vol_portfolio.diff()) & (ann_vol_portfolio.diff() > 10))).sum()
count_df_5 = ((ann_vol_portfolio  >=  27) | (ann_vol_portfolio.diff() > 25)).sum()

The exihibit below shows number of times volatility has breached the above condtions in the given investment horizon.

plt.figure(figsize=(15, 20))
plt.subplot(3,2,1)
count_df_1.plot.bar()
plt.ylabel('Count')
plt.title('CASE 1')

plt.subplot(3,2,2)
count_df_2.plot.bar()
plt.ylabel('Count')
plt.title('CASE 2')

plt.subplot(3,2,3)
count_df_3.plot.bar()
plt.ylabel('Count')
plt.title('CASE 3')

plt.subplot(3,2,4)
count_df_4.plot.bar()
plt.ylabel('Count')
plt.title('CASE 4')

plt.subplot(3,2,5)
count_df_5.plot.bar()
plt.ylabel('Count')
plt.title('CASE 5')

plt.show()

ann_vol_portfolio=ann_vol_portfolio.reset_index()
return_multiData=return_multiData.reset_index()

For ^FTSE

filtered_dates1 = ann_vol_portfolio[((ann_vol_portfolio['^FTSE'] < 15) & (ann_vol_portfolio['^FTSE'].diff() <10))]['DateTime']
filtered_dates4  = ann_vol_portfolio[(((27 > ann_vol_portfolio['^FTSE']) & (ann_vol_portfolio['^FTSE'] >= 25)) | ((15 > ann_vol_portfolio['^FTSE'].diff()) & (ann_vol_portfolio['^FTSE'].diff() > 10)))]['DateTime']
filtered_dates3  = ann_vol_portfolio[(((25 > ann_vol_portfolio['^FTSE']) & (ann_vol_portfolio['^FTSE'] >= 20)) | ((30 > ann_vol_portfolio['^FTSE'].diff()) & (ann_vol_portfolio['^FTSE'].diff() > 20)))]['DateTime']
filtered_dates2  = ann_vol_portfolio[(((20 > ann_vol_portfolio['^FTSE']) & (ann_vol_portfolio['^FTSE'] >= 15)) | ((20 > ann_vol_portfolio['^FTSE'].diff()) & (ann_vol_portfolio['^FTSE'].diff() > 15)))]['DateTime']
filtered_dates5  = ann_vol_portfolio[((ann_vol_portfolio['^FTSE']  >=  27) | (ann_vol_portfolio['^FTSE'].diff() >25))]['DateTime']

# Dates fullfilling the condition
def selected_returns_dataframe(filtered_dates):
  selected_returns = []
  for date in filtered_dates:
    return_value = return_multiData.loc[return_multiData['DateTime'] == date, '^FTSE'].values
    if len(return_value) > 0:
        selected_returns.append((date, return_value[0]))

  selected_return=pd.DataFrame(selected_returns)
  selected_return.set_axis(['Date', 'Portfolio_Return_ftse'], axis=1, inplace=True)
  return selected_return

def filtered_return_dataframe(selected_return_df):
  filtered_return_df = return_multiData[~return_multiData['DateTime'].isin(selected_return_df['Date'])]
  filtered_return_df1=pd.DataFrame()
  filtered_return_df1=filtered_return_df[['DateTime','^FTSE']]
  return filtered_return_df1

CASE 1

case1_ftse=selected_returns_dataframe(filtered_dates1)
case1_ftse.head()

	Date	Portfolio_Return_ftse
0	2013-02-22	0.007025
1	2013-02-25	0.003109
2	2013-03-01	0.002798
3	2013-03-04	-0.005174
4	2013-03-06	-0.000684

# Comparing filtered dates with the main return dataframes dates and extract date and return from return dataframe which is not fulfilling condition
filt_case1_ftse= filtered_return_dataframe(case1_ftse)
filt_case1_ftse.head()

	DateTime	^FTSE
0	2009-01-05	0.003902
1	2009-01-06	0.012949
2	2009-01-07	-0.028326
3	2009-01-09	-0.013089
4	2009-01-12	-0.005013

# cumulative product of return which is not satisfied the condition 1
filt_case1_ftse=filt_case1_ftse.set_index('DateTime')
filt_case1_ftse_cumpro=(1+filt_case1_ftse).cumprod()
filt_case1_ftse_cumpro.head()

	^FTSE
DateTime
2009-01-05	1.003902
2009-01-06	1.016901
2009-01-07	0.988097
2009-01-09	0.975163
2009-01-12	0.970275

case1_ftse.plot.hist()

<Axes: ylabel='Frequency'>

selected_return1_ftse=case1_ftse.set_index('Date')

def price_parameters(df_ser):
    return round(df_ser.mean(),2),round(df_ser.std(),2),round(df_ser.skew(),2),round(df_ser.kurtosis(),2)

#Plot distribution of series
def plot_dist(df):
    figure, axis = plt.subplots(1, 2)
    figure.set_figheight(10)
    figure.set_figwidth(20)
    mean,std,skew,kurt=price_parameters(df['Portfolio_Return_ftse']*100)
    sns.lineplot(data=df['Portfolio_Return_ftse']*100,ax=axis[0])
    axis[0].axhline(mean,color='r',ls='--')
    axis[0].set_title('Daily return chart(%), Total Returns:'+str(round(np.product(1+df['Portfolio_Return_ftse'])-1,2)*100)+'% Std-dev:'+str(std)+'%')
    sns.distplot(df['Portfolio_Return_ftse']*100,bins=50,ax=axis[1])
    x_axis=np.arange(min(df['Portfolio_Return_ftse']*100),max(df['Portfolio_Return_ftse']*100),0.0001)
    axis[1].axvline(mean,color='r',ls='--')
    axis[1].axvline(norm.ppf(0.05, loc=mean, scale=std),ls='--',color='r')
    axis[1].axvline(norm.ppf(0.95, loc=mean, scale=std),ls='--',color='r')
    axis[1].plot(x_axis,norm.pdf(x_axis, loc=mean, scale=std),ls='--',color='r')
    axis[1].set_title('Returns distribution, Mean:'+str(mean)+' std-dev:'+str(std)+' Skew:'+str(skew)+' Kurtosis:'+str(kurt))

plot_dist(selected_return1_ftse)

CASE 2

case2_ftse= selected_returns_dataframe(filtered_dates2)
case2_ftse.head()

	Date	Portfolio_Return_ftse
0	2010-04-20	0.009742
1	2010-04-21	-0.010426
2	2010-04-22	-0.010151
3	2010-04-23	0.010308
4	2010-04-27	-0.021000

filt_case2_ftse= filtered_return_dataframe(case2_ftse)
filt_case2_ftse.head()

	DateTime	^FTSE
0	2009-01-05	0.003902
1	2009-01-06	0.012949
2	2009-01-07	-0.028326
3	2009-01-09	-0.013089
4	2009-01-12	-0.005013

# cumulative product of return which is not satisfied the condition 2
filt_case2_ftse=filt_case2_ftse.set_index('DateTime')
filt_case2_ftse_cumpro=(1+filt_case2_ftse).cumprod()
filt_case2_ftse_cumpro.head()

	^FTSE
DateTime
2009-01-05	1.003902
2009-01-06	1.016901
2009-01-07	0.988097
2009-01-09	0.975163
2009-01-12	0.970275

# Histogram of ftse
case2_ftse.plot.hist()

<Axes: ylabel='Frequency'>

selected_return2_ftse=case2_ftse.set_index('Date')
plot_dist(selected_return2_ftse)

CASE 3

case3_ftse= selected_returns_dataframe(filtered_dates3)
case3_ftse.head()

	Date	Portfolio_Return_ftse
0	2010-02-10	0.003952
1	2010-02-11	0.005748
2	2010-02-17	0.022300
3	2010-02-18	0.009192
4	2010-02-19	0.006216

filt_case3_ftse= filtered_return_dataframe(case3_ftse)
filt_case3_ftse.head()

	DateTime	^FTSE
0	2009-01-05	0.003902
1	2009-01-06	0.012949
2	2009-01-07	-0.028326
3	2009-01-09	-0.013089
4	2009-01-12	-0.005013

# cumulative product of return which is not satisfied the condition 3
filt_case3_ftse=filt_case3_ftse.set_index('DateTime')
filt_case3_ftse_cumpro=(1+filt_case3_ftse).cumprod()
filt_case3_ftse_cumpro.head()

	^FTSE
DateTime
2009-01-05	1.003902
2009-01-06	1.016901
2009-01-07	0.988097
2009-01-09	0.975163
2009-01-12	0.970275

# Histogram of ftse
case3_ftse.plot.hist()

<Axes: ylabel='Frequency'>

selected_return3_ftse=case3_ftse.set_index('Date')
plot_dist(selected_return3_ftse)

CASE 4

case4_ftse= selected_returns_dataframe(filtered_dates4)
case4_ftse.head()

	Date	Portfolio_Return_ftse
0	2020-04-07	0.053365
1	2020-04-08	-0.004698
2	2020-04-09	0.029061
3	2020-04-15	-0.041933
4	2020-04-16	0.005484

filt_case4_ftse= filtered_return_dataframe(case4_ftse)
filt_case4_ftse.head()

	DateTime	^FTSE
0	2009-01-05	0.003902
1	2009-01-06	0.012949
2	2009-01-07	-0.028326
3	2009-01-09	-0.013089
4	2009-01-12	-0.005013

# cumulative product of return which is not satisfied the condition 4
filt_case4_ftse=filt_case4_ftse.set_index('DateTime')
filt_case4_ftse_cumpro=(1+filt_case4_ftse).cumprod()
filt_case4_ftse_cumpro.head()

	^FTSE
DateTime
2009-01-05	1.003902
2009-01-06	1.016901
2009-01-07	0.988097
2009-01-09	0.975163
2009-01-12	0.970275

# Histogram of ftse
case4_ftse.plot.hist()

<Axes: ylabel='Frequency'>

selected_return4_ftse=case4_ftse.set_index('Date')
plot_dist(selected_return4_ftse)

CASE 5

case5_ftse= selected_returns_dataframe(filtered_dates5)
case5_ftse.head()

	Date	Portfolio_Return_ftse
0	2020-05-04	-0.059114
1	2020-05-05	0.016615
2	2020-05-06	0.000752
3	2020-05-07	0.014042
4	2020-05-11	0.000623

filt_case5_ftse= filtered_return_dataframe(case5_ftse)
filt_case5_ftse.head()

	DateTime	^FTSE
0	2009-01-05	0.003902
1	2009-01-06	0.012949
2	2009-01-07	-0.028326
3	2009-01-09	-0.013089
4	2009-01-12	-0.005013

# cumulative product of return which is not satisfied the condition 5
filt_case5_ftse=filt_case5_ftse.set_index('DateTime')
filt_case5_ftse_cumpro=(1+filt_case5_ftse).cumprod()
filt_case5_ftse_cumpro.head()

	^FTSE
DateTime
2009-01-05	1.003902
2009-01-06	1.016901
2009-01-07	0.988097
2009-01-09	0.975163
2009-01-12	0.970275

# Histogram of ftse
case5_ftse.plot.hist()

<Axes: ylabel='Frequency'>

selected_return5_ftse=case5_ftse.set_index('Date')
plot_dist(selected_return5_ftse)

# Create a figure and axis
fig, ax = plt.subplots()

ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['^FTSE'], color='blue', label='^FTSE')


# Create a mask for the condition
mask1 = (ann_vol_portfolio['^FTSE'] < 15) & (ann_vol_portfolio['^FTSE'].diff() <10)

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['^FTSE'], where=mask1, color='lightgray',label='case 1')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('FTSE')
ax.legend(loc='upper right')

# Show the plot
plt.show()

# Create a figure and axis
fig, ax = plt.subplots()

ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['^FTSE'], color='yellow', label='FTSE')


# Create a mask for the condition
mask2 = (((27 > ann_vol_portfolio['^FTSE']) & (ann_vol_portfolio['^FTSE'] >= 25)) | ((15 > ann_vol_portfolio['^FTSE'].diff()) & (ann_vol_portfolio['^FTSE'].diff() > 10)))

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['^FTSE'], where=mask2, color='lightgray',label='case 2')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('FTSE')
ax.legend(loc='upper right')

# Show the plot
plt.show()

# Create a figure and axis
fig, ax = plt.subplots()

ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['^FTSE'], color='green', label='FTSE')


# Create a mask for the condition
mask3 = (((25 > ann_vol_portfolio['^FTSE']) & (ann_vol_portfolio['^FTSE'] >= 20)) | ((30 > ann_vol_portfolio['^FTSE'].diff()) & (ann_vol_portfolio['^FTSE'].diff() > 20)))

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['^FTSE'], where=mask3, color='lightgray',label='case 3')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('FTSE')
ax.legend(loc='upper right')

# Show the plot
plt.show()

# Create a figure and axis
fig, ax = plt.subplots()

ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['^FTSE'], color='red', label='FTSE')


# Create a mask for the condition
mask4 = (((20 > ann_vol_portfolio['^FTSE']) & (ann_vol_portfolio['^FTSE'] >= 15)) | ((20 > ann_vol_portfolio['^FTSE'].diff()) & (ann_vol_portfolio['^FTSE'].diff() > 15)))

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['^FTSE'], where=mask4, color='lightgray',label='case 4')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('FTSE')
ax.legend(loc='upper right')

# Show the plot
plt.show()

# Create a figure and axis
fig, ax = plt.subplots()

ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['^FTSE'], color='orange', label='FTSE')


# Create a mask for the condition
mask5 = (ann_vol_portfolio['^FTSE'] >= 27) | (ann_vol_portfolio['^FTSE'].diff() > 25)

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['^FTSE'], where=mask5, color='lightgray',label='case 5')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('FTSE')
ax.legend(loc='upper right')

# Show the plot
plt.show()

FOR ^GSPC

#Plot distribution of series
def plot_dist(df):
    figure, axis = plt.subplots(1, 2)
    figure.set_figheight(10)
    figure.set_figwidth(20)
    mean,std,skew,kurt=price_parameters(df['Portfolio_Return_gspc']*100)
    sns.lineplot(data=df['Portfolio_Return_gspc']*100,ax=axis[0])
    axis[0].axhline(mean,color='r',ls='--')
    axis[0].set_title('Daily return chart(%), Total Returns:'+str(round(np.product(1+df['Portfolio_Return_gspc'])-1,2)*100)+'% Std-dev:'+str(std)+'%')
    sns.distplot(df['Portfolio_Return_gspc']*100,bins=50,ax=axis[1])
    x_axis=np.arange(min(df['Portfolio_Return_gspc']*100),max(df['Portfolio_Return_gspc']*100),0.0001)
    axis[1].axvline(mean,color='r',ls='--')
    axis[1].axvline(norm.ppf(0.05, loc=mean, scale=std),ls='--',color='r')
    axis[1].axvline(norm.ppf(0.95, loc=mean, scale=std),ls='--',color='r')
    axis[1].plot(x_axis,norm.pdf(x_axis, loc=mean, scale=std),ls='--',color='r')
    axis[1].set_title('Returns distribution, Mean:'+str(mean)+' std-dev:'+str(std)+' Skew:'+str(skew)+' Kurtosis:'+str(kurt))

filtered_dates1 = ann_vol_portfolio[((ann_vol_portfolio['^GSPC'] < 15) & (ann_vol_portfolio['^GSPC'].diff() <10))]['DateTime']
filtered_dates4  = ann_vol_portfolio[(((27 > ann_vol_portfolio['^GSPC']) & (ann_vol_portfolio['^GSPC'] >= 25)) | ((15 > ann_vol_portfolio['^GSPC'].diff()) & (ann_vol_portfolio['^GSPC'].diff() > 10)))]['DateTime']
filtered_dates3  = ann_vol_portfolio[(((25 > ann_vol_portfolio['^GSPC']) & (ann_vol_portfolio['^GSPC'] >= 20)) | ((30 > ann_vol_portfolio['^GSPC'].diff()) & (ann_vol_portfolio['^GSPC'].diff() > 20)))]['DateTime']
filtered_dates2  = ann_vol_portfolio[(((20 > ann_vol_portfolio['^GSPC']) & (ann_vol_portfolio['^GSPC'] >= 15)) | ((20 > ann_vol_portfolio['^GSPC'].diff()) & (ann_vol_portfolio['^GSPC'].diff() > 15)))]['DateTime']
filtered_dates5  = ann_vol_portfolio[((ann_vol_portfolio['^GSPC']  >=  27) | (ann_vol_portfolio['^GSPC'].diff() >25))]['DateTime']

# Dates fullfilling the condition
def selected_returns_dataframe(filtered_dates):
  selected_returns = []
  for date in filtered_dates:
    return_value = return_multiData.loc[return_multiData['DateTime'] == date, '^GSPC'].values
    if len(return_value) > 0:
        selected_returns.append((date, return_value[0]))

  selected_return=pd.DataFrame(selected_returns)
  selected_return.set_axis(['Date', 'Portfolio_Return_gspc'], axis=1, inplace=True)
  return selected_return

def filtered_return_dataframe(selected_return_df):
  filtered_return_df = return_multiData[~return_multiData['DateTime'].isin(selected_return_df['Date'])]
  filtered_return_df1=pd.DataFrame()
  filtered_return_df1=filtered_return_df[['DateTime','^GSPC']]
  return filtered_return_df1

CASE 1

case1_gspc=selected_returns_dataframe(filtered_dates1)
case1_gspc.head()

	Date	Portfolio_Return_gspc
0	2013-01-23	0.001507
1	2013-01-24	0.000007
2	2013-01-25	0.005445
3	2013-01-29	0.003247
4	2013-01-30	-0.003900

# Comparing filtered dates with the main return dataframes dates and extract date and return from return dataframe which is not fulfilling condition

filt_case1_gspc= filtered_return_dataframe(case1_gspc)
filt_case1_gspc.head()

	DateTime	^GSPC
0	2009-01-05	-0.004668
1	2009-01-06	0.007817
2	2009-01-07	-0.030010
3	2009-01-09	-0.017978
4	2009-01-12	-0.022564

# cumulative product of return which is not satisfied the condition 1
filt_case1_gspc=filt_case1_gspc.set_index('DateTime')
filt_case1_gspc_cumpro=(1+filt_case1_gspc).cumprod()
filt_case1_gspc_cumpro.head()

	^GSPC
DateTime
2009-01-05	0.995332
2009-01-06	1.003112
2009-01-07	0.973009
2009-01-09	0.955516
2009-01-12	0.933956

case1_gspc.plot.hist()

<Axes: ylabel='Frequency'>

selected_return1_gspc=case1_gspc.set_index('Date')
plot_dist(selected_return1_gspc)

CASE 2

case2_gspc= selected_returns_dataframe(filtered_dates2)
case2_gspc.head()

	Date	Portfolio_Return_gspc
0	2010-06-23	-0.002985
1	2010-06-28	-0.002034
2	2010-07-02	-0.007888
3	2010-07-06	0.005359
4	2010-07-07	0.031331

filt_case2_gspc= filtered_return_dataframe(case2_gspc)
filt_case2_gspc.head()

	DateTime	^GSPC
0	2009-01-05	-0.004668
1	2009-01-06	0.007817
2	2009-01-07	-0.030010
3	2009-01-09	-0.017978
4	2009-01-12	-0.022564

# cumulative product of return which is not satisfied the condition 2
filt_case2_gspc=filt_case2_gspc.set_index('DateTime')
filt_case2_gspc_cumpro=(1+filt_case2_gspc).cumprod()
filt_case2_gspc_cumpro.head()

	^GSPC
DateTime
2009-01-05	0.995332
2009-01-06	1.003112
2009-01-07	0.973009
2009-01-09	0.955516
2009-01-12	0.933956

# Histogram of gspc
case2_gspc.plot.hist()

<Axes: ylabel='Frequency'>

selected_return2_gspc=case2_gspc.set_index('Date')
plot_dist(selected_return2_gspc)

CASE 3

case3_gspc= selected_returns_dataframe(filtered_dates3)
case3_gspc.head()

	Date	Portfolio_Return_gspc
0	2010-04-09	0.006684
1	2010-04-12	0.001767
2	2010-04-13	0.000685
3	2010-04-15	0.012002
4	2010-04-16	-0.016127

filt_case3_gspc= filtered_return_dataframe(case3_gspc)
filt_case3_gspc.head()

	DateTime	^GSPC
0	2009-01-05	-0.004668
1	2009-01-06	0.007817
2	2009-01-07	-0.030010
3	2009-01-09	-0.017978
4	2009-01-12	-0.022564

# cumulative product of return which is not satisfied the condition 3
filt_case3_gspc=filt_case3_gspc.set_index('DateTime')
filt_case3_gspc_cumpro=(1+filt_case3_gspc).cumprod()
filt_case3_gspc_cumpro.head()

	^GSPC
DateTime
2009-01-05	0.995332
2009-01-06	1.003112
2009-01-07	0.973009
2009-01-09	0.955516
2009-01-12	0.933956

# Histogram of gspc
case3_gspc.plot.hist()

<Axes: ylabel='Frequency'>

selected_return3_gspc=case3_gspc.set_index('Date')
plot_dist(selected_return3_gspc)

CASE 4

case4_gspc= selected_returns_dataframe(filtered_dates4)
case4_gspc.head()

	Date	Portfolio_Return_gspc
0	2010-03-03	0.000429
1	2010-03-04	0.003736
2	2010-03-05	0.014007
3	2010-03-08	-0.000176
4	2010-03-09	0.001713

filt_case4_gspc= filtered_return_dataframe(case4_gspc)
filt_case4_gspc.head()

	DateTime	^GSPC
0	2009-01-05	-0.004668
1	2009-01-06	0.007817
2	2009-01-07	-0.030010
3	2009-01-09	-0.017978
4	2009-01-12	-0.022564

# cumulative product of return which is not satisfied the condition 4
filt_case4_gspc=filt_case4_gspc.set_index('DateTime')
filt_case4_gspc_cumpro=(1+filt_case4_gspc).cumprod()
filt_case4_gspc_cumpro.head()

	^GSPC
DateTime
2009-01-05	0.995332
2009-01-06	1.003112
2009-01-07	0.973009
2009-01-09	0.955516
2009-01-12	0.933956

# Histogram of gspc
case4_gspc.plot.hist()

<Axes: ylabel='Frequency'>

selected_return4_gspc=case4_gspc.set_index('Date')
plot_dist(selected_return4_gspc)

CASE 5

case5_gspc= selected_returns_dataframe(filtered_dates5)
case5_gspc.head()

	Date	Portfolio_Return_gspc
0	2010-02-10	-0.002233
1	2010-02-11	0.009680
2	2010-02-17	0.019509
3	2010-02-18	0.006585
4	2010-02-19	0.002187

filt_case5_gspc= filtered_return_dataframe(case5_gspc)
filt_case5_gspc.head()

	DateTime	^GSPC
0	2009-01-05	-0.004668
1	2009-01-06	0.007817
2	2009-01-07	-0.030010
3	2009-01-09	-0.017978
4	2009-01-12	-0.022564

# cumulative product of return which is not satisfied the condition 5
filt_case5_gspc=filt_case5_gspc.set_index('DateTime')
filt_case5_gspc_cumpro=(1+filt_case5_gspc).cumprod()
filt_case5_gspc_cumpro.head()

	^GSPC
DateTime
2009-01-05	0.995332
2009-01-06	1.003112
2009-01-07	0.973009
2009-01-09	0.955516
2009-01-12	0.933956

# Histogram of gspc
case5_gspc.plot.hist()

<Axes: ylabel='Frequency'>

selected_return5_gspc=case5_gspc.set_index('Date')
plot_dist(selected_return5_gspc)

# Create a figure and axis
fig, ax = plt.subplots()

# Plot the entire '^GSPC' column
ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['^GSPC'], color='blue', label='GSPC')


# Create a mask for the condition
mask1 = (ann_vol_portfolio['^GSPC'] < 15) & (ann_vol_portfolio['^GSPC'].diff() <10)

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['^GSPC'], where=mask1, color='lightgray',label='case 1')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('GSPC')
ax.legend(loc='upper right')

# Show the plot
plt.show()

# Create a figure and axis
fig, ax = plt.subplots()

# Plot the entire '^GSPC' column
ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['^GSPC'], color='yellow', label='GSPC')


# Create a mask for the condition
mask2 = (((27 > ann_vol_portfolio['^GSPC']) & (ann_vol_portfolio['^GSPC'] >= 25)) | ((15 > ann_vol_portfolio['^GSPC'].diff()) & (ann_vol_portfolio['^GSPC'].diff() > 10)))

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['^GSPC'], where=mask2, color='lightgray',label='case 2')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('GSPC')
ax.legend(loc='upper right')

# Show the plot
plt.show()

# Create a figure and axis
fig, ax = plt.subplots()

# Plot the entire '^GSPC' column
ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['^GSPC'], color='green', label='GSPC')


# Create a mask for the condition
mask3 = (((25 > ann_vol_portfolio['^GSPC']) & (ann_vol_portfolio['^GSPC'] >= 20)) | ((30 > ann_vol_portfolio['^GSPC'].diff()) & (ann_vol_portfolio['^GSPC'].diff() > 20)))

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['^GSPC'], where=mask3, color='lightgray',label='case 3')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('GSPC')
ax.legend(loc='upper right')

# Show the plot
plt.show()

# Create a figure and axis
fig, ax = plt.subplots()

# Plot the entire '^GSPC' column
ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['^GSPC'], color='red', label='GSPC')


# Create a mask for the condition
mask4 = (((20 > ann_vol_portfolio['^GSPC']) & (ann_vol_portfolio['^GSPC'] >= 15)) | ((20 > ann_vol_portfolio['^GSPC'].diff()) & (ann_vol_portfolio['^GSPC'].diff() > 15)))

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['^GSPC'], where=mask4, color='lightgray',label='case 4')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('GSPC')
ax.legend(loc='upper right')

# Show the plot
plt.show()

# Create a figure and axis
fig, ax = plt.subplots()

# Plot the entire '^GSPC' column
ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['^GSPC'], color='orange', label='GSPC')


# Create a mask for the condition
mask5 = (ann_vol_portfolio['^GSPC'] >= 27) | (ann_vol_portfolio['^GSPC'].diff() > 25)

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['^GSPC'], where=mask5, color='lightgray',label='case 5')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('GSPC')
ax.legend(loc='upper right')

# Show the plot
plt.show()

FOR ^HSI

#Plot distribution of series
def plot_dist(df):
    figure, axis = plt.subplots(1, 2)
    figure.set_figheight(10)
    figure.set_figwidth(20)
    mean,std,skew,kurt=price_parameters(df['Portfolio_Return_hsi']*100)
    sns.lineplot(data=df['Portfolio_Return_hsi']*100,ax=axis[0])
    axis[0].axhline(mean,color='r',ls='--')
    axis[0].set_title('Daily return chart(%), Total Returns:'+str(round(np.product(1+df['Portfolio_Return_hsi'])-1,2)*100)+'% Std-dev:'+str(std)+'%')
    sns.distplot(df['Portfolio_Return_hsi']*100,bins=50,ax=axis[1])
    x_axis=np.arange(min(df['Portfolio_Return_hsi']*100),max(df['Portfolio_Return_hsi']*100),0.0001)
    axis[1].axvline(mean,color='r',ls='--')
    axis[1].axvline(norm.ppf(0.05, loc=mean, scale=std),ls='--',color='r')
    axis[1].axvline(norm.ppf(0.95, loc=mean, scale=std),ls='--',color='r')
    axis[1].plot(x_axis,norm.pdf(x_axis, loc=mean, scale=std),ls='--',color='r')
    axis[1].set_title('Returns distribution, Mean:'+str(mean)+' std-dev:'+str(std)+' Skew:'+str(skew)+' Kurtosis:'+str(kurt))

filtered_dates1 = ann_vol_portfolio[((ann_vol_portfolio['^HSI'] < 15) & (ann_vol_portfolio['^HSI'].diff() <10))]['DateTime']
filtered_dates4  = ann_vol_portfolio[(((27 > ann_vol_portfolio['^HSI']) & (ann_vol_portfolio['^HSI'] >= 25)) | ((15 > ann_vol_portfolio['^HSI'].diff()) & (ann_vol_portfolio['^HSI'].diff() > 10)))]['DateTime']
filtered_dates3  = ann_vol_portfolio[(((25 > ann_vol_portfolio['^HSI']) & (ann_vol_portfolio['^HSI'] >= 20)) | ((30 > ann_vol_portfolio['^HSI'].diff()) & (ann_vol_portfolio['^HSI'].diff() > 20)))]['DateTime']
filtered_dates2  = ann_vol_portfolio[(((20 > ann_vol_portfolio['^HSI']) & (ann_vol_portfolio['^HSI'] >= 15)) | ((20 > ann_vol_portfolio['^HSI'].diff()) & (ann_vol_portfolio['^HSI'].diff() > 15)))]['DateTime']
filtered_dates5  = ann_vol_portfolio[((ann_vol_portfolio['^HSI']  >=  27) | (ann_vol_portfolio['^HSI'].diff() >25))]['DateTime']

# Dates fullfilling the condition
def selected_returns_dataframe(filtered_dates):
  selected_returns = []
  for date in filtered_dates:
    return_value = return_multiData.loc[return_multiData['DateTime'] == date, '^HSI'].values
    if len(return_value) > 0:
        selected_returns.append((date, return_value[0]))

  selected_return=pd.DataFrame(selected_returns)
  selected_return.set_axis(['Date', 'Portfolio_Return_hsi'], axis=1, inplace=True)
  return selected_return

def filtered_return_dataframe(selected_return_df):
  filtered_return_df = return_multiData[~return_multiData['DateTime'].isin(selected_return_df['Date'])]
  filtered_return_df1=pd.DataFrame()
  filtered_return_df1=filtered_return_df[['DateTime','^HSI']]
  return filtered_return_df1

CASE 1

case1_hsi=selected_returns_dataframe(filtered_dates1)
case1_hsi.head()

	Date	Portfolio_Return_hsi
0	2015-03-20	-0.003827
1	2015-03-23	0.004893
2	2015-03-24	-0.003875
3	2015-03-25	0.005272
4	2015-03-26	-0.001270

# Comparing filtered dates with the main return dataframes dates and extract date and return from return dataframe which is not fulfilling condition

filt_case1_hsi= filtered_return_dataframe(case1_hsi)
filt_case1_hsi.head()

	DateTime	^HSI
0	2009-01-05	0.034601
1	2009-01-06	-0.003457
2	2009-01-07	-0.033660
3	2009-01-09	-0.040702
4	2009-01-12	-0.028269

# cumulative product of return which is not satisfied the condition 1
filt_case1_hsi=filt_case1_hsi.set_index('DateTime')
filt_case1_hsi_cumpro=(1+filt_case1_hsi).cumprod()
filt_case1_hsi_cumpro.head()

	^HSI
DateTime
2009-01-05	1.034601
2009-01-06	1.031025
2009-01-07	0.996321
2009-01-09	0.955768
2009-01-12	0.928749

case1_hsi.plot.hist()

<Axes: ylabel='Frequency'>

selected_return1_hsi=case1_hsi.set_index('Date')
plot_dist(selected_return1_hsi)

CASE 2

case2_hsi= selected_returns_dataframe(filtered_dates2)
case2_hsi.head()

	Date	Portfolio_Return_hsi
0	2010-12-31	0.001570
1	2011-01-04	0.027481
2	2011-01-05	0.003775
3	2011-01-06	0.001199
4	2011-01-07	-0.004190

filt_case2_hsi= filtered_return_dataframe(case2_hsi)
filt_case2_hsi.head()

	DateTime	^HSI
0	2009-01-05	0.034601
1	2009-01-06	-0.003457
2	2009-01-07	-0.033660
3	2009-01-09	-0.040702
4	2009-01-12	-0.028269

# cumulative product of return which is not satisfied the condition 2
filt_case2_hsi=filt_case2_hsi.set_index('DateTime')
filt_case2_hsi_cumpro=(1+filt_case2_hsi).cumprod()
filt_case2_hsi_cumpro.head()

	^HSI
DateTime
2009-01-05	1.034601
2009-01-06	1.031025
2009-01-07	0.996321
2009-01-09	0.955768
2009-01-12	0.928749

# Histogram of hsi
case2_hsi.plot.hist()

<Axes: ylabel='Frequency'>

selected_return2_hsi=case2_hsi.set_index('Date')
plot_dist(selected_return2_hsi)

CASE 3

case3_hsi= selected_returns_dataframe(filtered_dates3)
case3_hsi.head()

	Date	Portfolio_Return_hsi
0	2010-07-07	-0.011305
1	2010-07-08	0.009744
2	2010-07-09	0.016364
3	2010-07-12	0.004356
4	2010-07-13	-0.001777

filt_case3_hsi= filtered_return_dataframe(case3_hsi)
filt_case3_hsi.head()

	DateTime	^HSI
0	2009-01-05	0.034601
1	2009-01-06	-0.003457
2	2009-01-07	-0.033660
3	2009-01-09	-0.040702
4	2009-01-12	-0.028269

# cumulative product of return which is not satisfied the condition 3
filt_case3_hsi=filt_case3_hsi.set_index('DateTime')
filt_case3_hsi_cumpro=(1+filt_case3_hsi).cumprod()
filt_case3_hsi_cumpro.head()

	^HSI
DateTime
2009-01-05	1.034601
2009-01-06	1.031025
2009-01-07	0.996321
2009-01-09	0.955768
2009-01-12	0.928749

# Histogram of hsi
case3_hsi.plot.hist()

<Axes: ylabel='Frequency'>

selected_return3_hsi=case3_hsi.set_index('Date')
plot_dist(selected_return3_hsi)

CASE 4

case4_hsi= selected_returns_dataframe(filtered_dates4)
case4_hsi.head()

	Date	Portfolio_Return_hsi
0	2010-06-07	-0.020319
1	2010-06-08	0.005642
2	2010-06-09	0.006864
3	2010-06-10	0.000584
4	2010-06-11	0.012208

filt_case4_hsi= filtered_return_dataframe(case4_hsi)
filt_case4_hsi.head()

	DateTime	^HSI
0	2009-01-05	0.034601
1	2009-01-06	-0.003457
2	2009-01-07	-0.033660
3	2009-01-09	-0.040702
4	2009-01-12	-0.028269

# cumulative product of return which is not satisfied the condition 4
filt_case4_hsi=filt_case4_hsi.set_index('DateTime')
filt_case4_hsi_cumpro=(1+filt_case4_hsi).cumprod()
filt_case4_hsi_cumpro.head()

	^HSI
DateTime
2009-01-05	1.034601
2009-01-06	1.031025
2009-01-07	0.996321
2009-01-09	0.955768
2009-01-12	0.928749

# Histogram of hsi
case4_hsi.plot.hist()

<Axes: ylabel='Frequency'>

selected_return4_hsi=case4_hsi.set_index('Date')
plot_dist(selected_return4_hsi)

CASE 5

case5_hsi= selected_returns_dataframe(filtered_dates5)
case5_hsi.head()

	Date	Portfolio_Return_hsi
0	2010-02-10	0.006667
1	2010-02-11	0.018495
2	2010-02-17	0.011992
3	2010-02-18	-0.005448
4	2010-02-19	-0.025861

filt_case5_hsi= filtered_return_dataframe(case5_hsi)
filt_case5_hsi.head()

	DateTime	^HSI
0	2009-01-05	0.034601
1	2009-01-06	-0.003457
2	2009-01-07	-0.033660
3	2009-01-09	-0.040702
4	2009-01-12	-0.028269

# cumulative product of return which is not satisfied the condition 5
filt_case5_hsi=filt_case5_hsi.set_index('DateTime')
filt_case5_hsi_cumpro=(1+filt_case5_hsi).cumprod()
filt_case5_hsi_cumpro.head()

	^HSI
DateTime
2009-01-05	1.034601
2009-01-06	1.031025
2009-01-07	0.996321
2009-01-09	0.955768
2009-01-12	0.928749

# Histogram of hsi
case5_hsi.plot.hist()

<Axes: ylabel='Frequency'>

selected_return5_hsi=case5_hsi.set_index('Date')
plot_dist(selected_return5_hsi)

# Create a figure and axis
fig, ax = plt.subplots()

ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['^HSI'], color='blue', label='HSI')


# Create a mask for the condition
mask1 = (ann_vol_portfolio['^HSI'] < 15) & (ann_vol_portfolio['^HSI'].diff() <10)

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['^HSI'], where=mask1, color='lightgray',label='case 1')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('HSI')
ax.legend(loc='upper right')

# Show the plot
plt.show()

# Create a figure and axis
fig, ax = plt.subplots()

ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['^HSI'], color='yellow', label='HSI')


# Create a mask for the condition
mask2 = (((27 > ann_vol_portfolio['^HSI']) & (ann_vol_portfolio['^HSI'] >= 25)) | ((15 > ann_vol_portfolio['^HSI'].diff()) & (ann_vol_portfolio['^HSI'].diff() > 10)))

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['^HSI'], where=mask2, color='lightgray',label='case 2')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('HSI')
ax.legend(loc='upper right')

# Show the plot
plt.show()

# Create a figure and axis
fig, ax = plt.subplots()

ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['^HSI'], color='green', label='HSI')


# Create a mask for the condition
mask3 = (((25 > ann_vol_portfolio['^HSI']) & (ann_vol_portfolio['^HSI'] >= 20)) | ((30 > ann_vol_portfolio['^HSI'].diff()) & (ann_vol_portfolio['^HSI'].diff() > 20)))

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['^HSI'], where=mask3, color='lightgray',label='case 3')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('HSI')
ax.legend(loc='upper right')

# Show the plot
plt.show()

# Create a figure and axis
fig, ax = plt.subplots()

ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['^HSI'], color='red', label='HSI')


# Create a mask for the condition
mask4 = (((20 > ann_vol_portfolio['^HSI']) & (ann_vol_portfolio['^HSI'] >= 15)) | ((20 > ann_vol_portfolio['^HSI'].diff()) & (ann_vol_portfolio['^HSI'].diff() > 15)))

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['^HSI'], where=mask4, color='lightgray',label='case 4')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('HSI')
ax.legend(loc='upper right')

# Show the plot
plt.show()

# Create a figure and axis
fig, ax = plt.subplots()

ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['^HSI'], color='orange', label='HSI')


# Create a mask for the condition
mask5 = (ann_vol_portfolio['^HSI'] >= 27) | (ann_vol_portfolio['^HSI'].diff() > 25)

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['^HSI'], where=mask5, color='lightgray',label='case 5')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('HSI')
ax.legend(loc='upper right')

# Show the plot
plt.show()

FOR ^NSEI

#Plot distribution of series
def plot_dist(df):
    figure, axis = plt.subplots(1, 2)
    figure.set_figheight(10)
    figure.set_figwidth(20)
    mean,std,skew,kurt=price_parameters(df['Portfolio_Return_nsei']*100)
    sns.lineplot(data=df['Portfolio_Return_nsei']*100,ax=axis[0])
    axis[0].axhline(mean,color='r',ls='--')
    axis[0].set_title('Daily return chart(%), Total Returns:'+str(round(np.product(1+df['Portfolio_Return_nsei'])-1,2)*100)+'% Std-dev:'+str(std)+'%')
    sns.distplot(df['Portfolio_Return_nsei']*100,bins=50,ax=axis[1])
    x_axis=np.arange(min(df['Portfolio_Return_nsei']*100),max(df['Portfolio_Return_nsei']*100),0.0001)
    axis[1].axvline(mean,color='r',ls='--')
    axis[1].axvline(norm.ppf(0.05, loc=mean, scale=std),ls='--',color='r')
    axis[1].axvline(norm.ppf(0.95, loc=mean, scale=std),ls='--',color='r')
    axis[1].plot(x_axis,norm.pdf(x_axis, loc=mean, scale=std),ls='--',color='r')
    axis[1].set_title('Returns distribution, Mean:'+str(mean)+' std-dev:'+str(std)+' Skew:'+str(skew)+' Kurtosis:'+str(kurt))

filtered_dates1 = ann_vol_portfolio[((ann_vol_portfolio['^NSEI'] < 15) & (ann_vol_portfolio['^NSEI'].diff() <10))]['DateTime']
filtered_dates4  = ann_vol_portfolio[(((27 > ann_vol_portfolio['^NSEI']) & (ann_vol_portfolio['^NSEI'] >= 25)) | ((15 > ann_vol_portfolio['^NSEI'].diff()) & (ann_vol_portfolio['^NSEI'].diff() > 10)))]['DateTime']
filtered_dates3  = ann_vol_portfolio[(((25 > ann_vol_portfolio['^NSEI']) & (ann_vol_portfolio['^NSEI'] >= 20)) | ((30 > ann_vol_portfolio['^NSEI'].diff()) & (ann_vol_portfolio['^NSEI'].diff() > 20)))]['DateTime']
filtered_dates2  = ann_vol_portfolio[(((20 > ann_vol_portfolio['^NSEI']) & (ann_vol_portfolio['^NSEI'] >= 15)) | ((20 > ann_vol_portfolio['^NSEI'].diff()) & (ann_vol_portfolio['^NSEI'].diff() > 15)))]['DateTime']
filtered_dates5  = ann_vol_portfolio[((ann_vol_portfolio['^NSEI']  >=  27) | (ann_vol_portfolio['^NSEI'].diff() >25))]['DateTime']

# Dates fullfilling the condition
def selected_returns_dataframe(filtered_dates):
  selected_returns = []
  for date in filtered_dates:
    return_value = return_multiData.loc[return_multiData['DateTime'] == date, '^NSEI'].values
    if len(return_value) > 0:
        selected_returns.append((date, return_value[0]))

  selected_return=pd.DataFrame(selected_returns)
  selected_return.set_axis(['Date', 'Portfolio_Return_nsei'], axis=1, inplace=True)
  return selected_return

def filtered_return_dataframe(selected_return_df):
  filtered_return_df = return_multiData[~return_multiData['DateTime'].isin(selected_return_df['Date'])]
  filtered_return_df1=pd.DataFrame()
  filtered_return_df1=filtered_return_df[['DateTime','^NSEI']]
  return filtered_return_df1

CASE 1

case1_nsei=selected_returns_dataframe(filtered_dates1)
case1_nsei.head()

	Date	Portfolio_Return_nsei
0	2014-11-11	0.002205
1	2014-11-12	0.002469
2	2014-11-13	-0.003036
3	2014-11-14	0.003835
4	2014-11-17	0.004869

# Comparing filtered dates with the main return dataframes dates and extract date and return from return dataframe which is not fulfilling condition

filt_case1_nsei= filtered_return_dataframe(case1_nsei)
filt_case1_nsei.head()

	DateTime	^NSEI
0	2009-01-05	0.024518
1	2009-01-06	-0.002771
2	2009-01-07	-0.061809
3	2009-01-09	-0.016231
4	2009-01-12	-0.034772

# cumulative product of return which is not satisfied the condition 1
filt_case1_nsei=filt_case1_nsei.set_index('DateTime')
filt_case1_nsei_cumpro=(1+filt_case1_nsei).cumprod()
filt_case1_nsei_cumpro.head()

	^NSEI
DateTime
2009-01-05	1.024518
2009-01-06	1.021679
2009-01-07	0.958530
2009-01-09	0.942972
2009-01-12	0.910183

case1_nsei.plot.hist()

<Axes: ylabel='Frequency'>

selected_return1_nsei=case1_nsei.set_index('Date')
plot_dist(selected_return1_nsei)

CASE 2

case2_nsei= selected_returns_dataframe(filtered_dates2)
case2_nsei.head()

	Date	Portfolio_Return_nsei
0	2010-08-20	-0.001724
1	2010-08-23	0.002323
2	2010-08-24	-0.006927
3	2010-08-25	-0.007766
4	2010-08-26	0.002847

filt_case2_nsei= filtered_return_dataframe(case2_nsei)
filt_case2_nsei.head()

	DateTime	^NSEI
0	2009-01-05	0.024518
1	2009-01-06	-0.002771
2	2009-01-07	-0.061809
3	2009-01-09	-0.016231
4	2009-01-12	-0.034772

# cumulative product of return which is not satisfied the condition 2
filt_case2_nsei=filt_case2_nsei.set_index('DateTime')
filt_case2_nsei_cumpro=(1+filt_case2_nsei).cumprod()
filt_case2_nsei_cumpro.head()

	^NSEI
DateTime
2009-01-05	1.024518
2009-01-06	1.021679
2009-01-07	0.958530
2009-01-09	0.942972
2009-01-12	0.910183

# Histogram of nsei
case2_nsei.plot.hist()

<Axes: ylabel='Frequency'>

selected_return2_nsei=case2_nsei.set_index('Date')
plot_dist(selected_return2_nsei)

CASE 3

case3_nsei= selected_returns_dataframe(filtered_dates3)
case3_nsei.head()

	Date	Portfolio_Return_nsei
0	2010-06-23	0.001241
1	2010-06-24	-0.000479
2	2010-06-25	-0.009689
3	2010-06-28	0.012232
4	2010-06-29	-0.014503

filt_case3_nsei= filtered_return_dataframe(case3_nsei)
filt_case3_nsei.head()

	DateTime	^NSEI
0	2009-01-05	0.024518
1	2009-01-06	-0.002771
2	2009-01-07	-0.061809
3	2009-01-09	-0.016231
4	2009-01-12	-0.034772

# cumulative product of return which is not satisfied the condition 3
filt_case3_nsei=filt_case3_nsei.set_index('DateTime')
filt_case3_nsei_cumpro=(1+filt_case3_nsei).cumprod()
filt_case3_nsei_cumpro.head()

	^NSEI
DateTime
2009-01-05	1.024518
2009-01-06	1.021679
2009-01-07	0.958530
2009-01-09	0.942972
2009-01-12	0.910183

# Histogram of nsei
case3_nsei.plot.hist()

<Axes: ylabel='Frequency'>

selected_return3_nsei=case3_nsei.set_index('Date')
plot_dist(selected_return3_nsei)

CASE 4

case4_nsei= selected_returns_dataframe(filtered_dates4)
case4_nsei.head()

	Date	Portfolio_Return_nsei
0	2020-03-23	-0.129805
1	2020-03-24	0.025071
2	2020-03-25	0.066247

filt_case4_nsei= filtered_return_dataframe(case4_nsei)
filt_case4_nsei.head()

	DateTime	^NSEI
0	2009-01-05	0.024518
1	2009-01-06	-0.002771
2	2009-01-07	-0.061809
3	2009-01-09	-0.016231
4	2009-01-12	-0.034772

# cumulative product of return which is not satisfied the condition 4
filt_case4_nsei=filt_case4_nsei.set_index('DateTime')
filt_case4_nsei_cumpro=(1+filt_case4_nsei).cumprod()
filt_case4_nsei_cumpro.head()

	^NSEI
DateTime
2009-01-05	1.024518
2009-01-06	1.021679
2009-01-07	0.958530
2009-01-09	0.942972
2009-01-12	0.910183

# Histogram of nsei
case4_nsei.plot.hist()

<Axes: ylabel='Frequency'>

selected_return4_nsei=case4_nsei.set_index('Date')
plot_dist(selected_return4_nsei)

CASE 5

case5_nsei= selected_returns_dataframe(filtered_dates5)
case5_nsei.head()

	Date	Portfolio_Return_nsei
0	2010-02-10	-0.007397
1	2010-02-11	0.014641
2	2010-02-17	0.018055
3	2010-02-18	-0.005342
4	2010-02-19	-0.008767

filt_case5_nsei= filtered_return_dataframe(case5_nsei)
filt_case5_nsei.head()

	DateTime	^NSEI
0	2009-01-05	0.024518
1	2009-01-06	-0.002771
2	2009-01-07	-0.061809
3	2009-01-09	-0.016231
4	2009-01-12	-0.034772

# cumulative product of return which is not satisfied the condition 5
filt_case5_nsei=filt_case5_nsei.set_index('DateTime')
filt_case5_nsei_cumpro=(1+filt_case5_nsei).cumprod()
filt_case5_nsei_cumpro.head()

	^NSEI
DateTime
2009-01-05	1.024518
2009-01-06	1.021679
2009-01-07	0.958530
2009-01-09	0.942972
2009-01-12	0.910183

# Histogram of nsei
case5_nsei.plot.hist()

<Axes: ylabel='Frequency'>

selected_return5_nsei=case5_nsei.set_index('Date')
plot_dist(selected_return5_nsei)

# Create a figure and axis
fig, ax = plt.subplots()

ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['^NSEI'], color='blue', label='NSEI')


# Create a mask for the condition
mask1 = (ann_vol_portfolio['^NSEI'] < 15) & (ann_vol_portfolio['^NSEI'].diff() <10)

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['^NSEI'], where=mask1, color='lightgray',label='case 1')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('NSEI')
ax.legend(loc='upper right')

# Show the plot
plt.show()

# Create a figure and axis
fig, ax = plt.subplots()

ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['^NSEI'], color='yellow', label='NSEI')


# Create a mask for the condition
mask2 = (((27 > ann_vol_portfolio['^NSEI']) & (ann_vol_portfolio['^NSEI'] >= 25)) | ((15 > ann_vol_portfolio['^NSEI'].diff()) & (ann_vol_portfolio['^NSEI'].diff() > 10)))

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['^NSEI'], where=mask2, color='lightgray',label='case 2')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('NSEI')
ax.legend(loc='upper right')

# Show the plot
plt.show()

# Create a figure and axis
fig, ax = plt.subplots()

ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['^NSEI'], color='green', label='NSEI')


# Create a mask for the condition
mask3 = (((25 > ann_vol_portfolio['^NSEI']) & (ann_vol_portfolio['^NSEI'] >= 20)) | ((30 > ann_vol_portfolio['^NSEI'].diff()) & (ann_vol_portfolio['^NSEI'].diff() > 20)))

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['^NSEI'], where=mask3, color='lightgray',label='case 3')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('NSEI')
ax.legend(loc='upper right')

# Show the plot
plt.show()

# Create a figure and axis
fig, ax = plt.subplots()

ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['^NSEI'], color='red', label='NSEI')


# Create a mask for the condition
mask4 = (((20 > ann_vol_portfolio['^NSEI']) & (ann_vol_portfolio['^NSEI'] >= 15)) | ((20 > ann_vol_portfolio['^NSEI'].diff()) & (ann_vol_portfolio['^NSEI'].diff() > 15)))

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['^NSEI'], where=mask4, color='lightgray',label='case 4')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('NSEI')
ax.legend(loc='upper right')

# Show the plot
plt.show()

# Create a figure and axis
fig, ax = plt.subplots()

ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['^NSEI'], color='orange', label='NSEI')


# Create a mask for the condition
mask5 = (ann_vol_portfolio['^NSEI'] >= 27) | (ann_vol_portfolio['^NSEI'].diff() > 25)

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['^NSEI'], where=mask5, color='lightgray',label='case 5')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('NSEI')
ax.legend(loc='upper right')

# Show the plot
plt.show()

FOR GLD

#Plot distribution of series
def plot_dist(df):
    figure, axis = plt.subplots(1, 2)
    figure.set_figheight(10)
    figure.set_figwidth(20)
    mean,std,skew,kurt=price_parameters(df['Portfolio_Return_gld']*100)
    sns.lineplot(data=df['Portfolio_Return_gld']*100,ax=axis[0])
    axis[0].axhline(mean,color='r',ls='--')
    axis[0].set_title('Daily return chart(%), Total Returns:'+str(round(np.product(1+df['Portfolio_Return_gld'])-1,2)*100)+'% Std-dev:'+str(std)+'%')
    sns.distplot(df['Portfolio_Return_gld']*100,bins=50,ax=axis[1])
    x_axis=np.arange(min(df['Portfolio_Return_gld']*100),max(df['Portfolio_Return_gld']*100),0.0001)
    axis[1].axvline(mean,color='r',ls='--')
    axis[1].axvline(norm.ppf(0.05, loc=mean, scale=std),ls='--',color='r')
    axis[1].axvline(norm.ppf(0.95, loc=mean, scale=std),ls='--',color='r')
    axis[1].plot(x_axis,norm.pdf(x_axis, loc=mean, scale=std),ls='--',color='r')
    axis[1].set_title('Returns distribution, Mean:'+str(mean)+' std-dev:'+str(std)+' Skew:'+str(skew)+' Kurtosis:'+str(kurt))

filtered_dates1 = ann_vol_portfolio[((ann_vol_portfolio['^GSPC'] < 15) & (ann_vol_portfolio['^GSPC'].diff() <10))]['DateTime']
filtered_dates4  = ann_vol_portfolio[(((27 > ann_vol_portfolio['^GSPC']) & (ann_vol_portfolio['^GSPC'] >= 25)) | ((15 > ann_vol_portfolio['^GSPC'].diff()) & (ann_vol_portfolio['^GSPC'].diff() > 10)))]['DateTime']
filtered_dates3  = ann_vol_portfolio[(((25 > ann_vol_portfolio['^GSPC']) & (ann_vol_portfolio['^GSPC'] >= 20)) | ((30 > ann_vol_portfolio['^GSPC'].diff()) & (ann_vol_portfolio['^GSPC'].diff() > 20)))]['DateTime']
filtered_dates2  = ann_vol_portfolio[(((20 > ann_vol_portfolio['^GSPC']) & (ann_vol_portfolio['^GSPC'] >= 15)) | ((20 > ann_vol_portfolio['^GSPC'].diff()) & (ann_vol_portfolio['^GSPC'].diff() > 15)))]['DateTime']
filtered_dates5  = ann_vol_portfolio[((ann_vol_portfolio['^GSPC']  >=  27) | (ann_vol_portfolio['^GSPC'].diff() >25))]['DateTime']

# Dates fullfilling the condition
def selected_returns_dataframe(filtered_dates):
  selected_returns = []
  for date in filtered_dates:
    return_value = return_multiData.loc[return_multiData['DateTime'] == date, 'GLD'].values
    if len(return_value) > 0:
        selected_returns.append((date, return_value[0]))

  selected_return=pd.DataFrame(selected_returns)
  selected_return.set_axis(['Date', 'Portfolio_Return_gld'], axis=1, inplace=True)
  return selected_return

CASE 1

case1_gld=selected_returns_dataframe(filtered_dates1)
case1_gld.head()

	Date	Portfolio_Return_gld
0	2013-01-23	-0.002810
1	2013-01-24	-0.010968
2	2013-01-25	-0.004770
3	2013-01-29	0.002116
4	2013-01-30	0.007454

case1_gld.plot.hist()

<Axes: ylabel='Frequency'>

selected_return1_gld=case1_gld.set_index('Date')
plot_dist(selected_return1_gld)

CASE 2

case2_gld= selected_returns_dataframe(filtered_dates2)
case2_gld.head()

	Date	Portfolio_Return_gld
0	2010-06-23	-0.004117
1	2010-06-28	-0.013604
2	2010-07-02	-0.026216
3	2010-07-06	-0.016710
4	2010-07-07	0.010471

# Histogram of gld
case2_gld.plot.hist()

<Axes: ylabel='Frequency'>

selected_return2_gld=case2_gld.set_index('Date')
plot_dist(selected_return2_gld)

CASE 3

case3_gld= selected_returns_dataframe(filtered_dates3)
case3_gld.head()

	Date	Portfolio_Return_gld
0	2010-04-09	0.008788
1	2010-04-12	-0.005544
2	2010-04-13	-0.002832
3	2010-04-15	0.008519
4	2010-04-16	-0.021205

# Histogram of gld
case3_gld.plot.hist()

<Axes: ylabel='Frequency'>

selected_return3_gld=case3_gld.set_index('Date')
plot_dist(selected_return3_gld)

CASE 4

case4_gld= selected_returns_dataframe(filtered_dates4)
case4_gld.head()

	Date	Portfolio_Return_gld
0	2010-03-03	0.005495
1	2010-03-04	-0.007166
2	2010-03-05	-0.000180
3	2010-03-08	-0.008393
4	2010-03-09	-0.001456

# Histogram of gld
case4_gld.plot.hist()

<Axes: ylabel='Frequency'>

selected_return4_gld=case4_gld.set_index('Date')
plot_dist(selected_return4_gld)

CASE 5

case5_gld= selected_returns_dataframe(filtered_dates5)
case5_gld.head()

	Date	Portfolio_Return_gld
0	2010-02-10	-0.002751
1	2010-02-11	0.019121
2	2010-02-17	0.019789
3	2010-02-18	0.006682
4	2010-02-19	-0.004637

# Histogram of gld
case5_gld.plot.hist()

<Axes: ylabel='Frequency'>

selected_return5_gld=case5_gld.set_index('Date')
plot_dist(selected_return5_gld)

# Create a figure and axis
fig, ax = plt.subplots()

ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['GLD'], color='blue', label='GLD')


# Create a mask for the condition
mask1 = (ann_vol_portfolio['^GSPC'] < 15) & (ann_vol_portfolio['^GSPC'].diff() <10)

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['GLD'], where=mask1, color='lightgray',label='case 1')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('GOLD')
ax.legend(loc='upper right')

# Show the plot
plt.show()

# Create a figure and axis
fig, ax = plt.subplots()

ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['GLD'], color='yellow', label='GLD')


# Create a mask for the condition
mask2 = (((27 > ann_vol_portfolio['^GSPC']) & (ann_vol_portfolio['^GSPC'] >= 25)) | ((15 > ann_vol_portfolio['^GSPC'].diff()) & (ann_vol_portfolio['^GSPC'].diff() > 10)))

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['GLD'], where=mask2, color='lightgray',label='case 2')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('GOLD')
ax.legend(loc='upper right')

# Show the plot
plt.show()

# Create a figure and axis
fig, ax = plt.subplots()

ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['GLD'], color='green', label='GLD')


# Create a mask for the condition
mask3 = (((25 > ann_vol_portfolio['^GSPC']) & (ann_vol_portfolio['^GSPC'] >= 20)) | ((30 > ann_vol_portfolio['^GSPC'].diff()) & (ann_vol_portfolio['^GSPC'].diff() > 20)))

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['GLD'], where=mask3, color='lightgray',label='case 3')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('GOLD')
ax.legend(loc='upper right')

# Show the plot
plt.show()

# Create a figure and axis
fig, ax = plt.subplots()

ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['GLD'], color='red', label='GLD')


# Create a mask for the condition
mask4 = (((20 > ann_vol_portfolio['^GSPC']) & (ann_vol_portfolio['^GSPC'] >= 15)) | ((20 > ann_vol_portfolio['^GSPC'].diff()) & (ann_vol_portfolio['^GSPC'].diff() > 15)))

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['GLD'], where=mask4, color='lightgray',label='case 4')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('GOLD')
ax.legend(loc='upper right')

# Show the plot
plt.show()

# Create a figure and axis
fig, ax = plt.subplots()

ax.plot(ann_vol_portfolio['DateTime'],ann_vol_portfolio['GLD'], color='orange', label='GLD')


# Create a mask for the condition
mask5 = (ann_vol_portfolio['^GSPC'] >= 27) | (ann_vol_portfolio['^GSPC'].diff() > 25)

# Shade the regions where the condition is met
ax.fill_between(ann_vol_portfolio['DateTime'], ann_vol_portfolio['GLD'], where=mask5, color='lightgray',label='case 5')


# Set labels and title
ax.set_xlabel('Date')
ax.set_ylabel('Volatility')
ax.set_title('GOLD')
ax.legend(loc='upper right')

# Show the plot
plt.show()

BOND DATA

from google.colab import files
uploaded = files.upload()

Upload widget is only available when the cell has been executed in the current browser session. Please rerun this cell to enable.

Saving China 10-Year Bond Yield Historical Data.csv to China 10-Year Bond Yield Historical Data (2).csv
Saving India 10-Year Bond Yield Historical Data.csv to India 10-Year Bond Yield Historical Data (2).csv
Saving United Kingdom 10-Year Bond Yield Historical Data.csv to United Kingdom 10-Year Bond Yield Historical Data (2).csv
Saving United States 10-Year Bond Yield Historical Data.csv to United States 10-Year Bond Yield Historical Data (2).csv

df_data_us=pd.read_csv('United States 10-Year Bond Yield Historical Data.csv')
df_data_ind=pd.read_csv('India 10-Year Bond Yield Historical Data.csv')
df_data_chi=pd.read_csv('China 10-Year Bond Yield Historical Data.csv')
df_data_uk=pd.read_csv('United Kingdom 10-Year Bond Yield Historical Data.csv')

df_data_us=df_data_us.set_index('Date')
df_data_chi=df_data_chi.set_index('Date')
df_data_ind=df_data_ind.set_index('Date')
df_data_uk=df_data_uk.set_index('Date')

df_bond=pd.DataFrame()
df_bond['US_Price']=df_data_us['Price']
df_bond['IN_Price']=df_data_ind['Price']
df_bond['UK_Price']=df_data_uk['Price']

df_bond['CHI_Price']=df_data_chi['Price']

df_bond.head()

	US_Price	IN_Price	UK_Price	CHI_Price
Date
01/05/2023	3.722	7.327	3.552	2.870
01/04/2023	3.690	7.321	3.493	2.863
01/03/2023	3.752	7.321	3.653	2.873
01/02/2023	3.879	7.340	3.699	NaN
12/30/2022	3.879	7.327	3.669	2.884

df_bond=df_bond.reset_index()
df_bond

	Date	US_Price	IN_Price	UK_Price	CHI_Price
0	01/05/2023	3.722	7.327	3.552	2.870
1	01/04/2023	3.690	7.321	3.493	2.863
2	01/03/2023	3.752	7.321	3.653	2.873
3	01/02/2023	3.879	7.340	3.699	NaN
4	12/30/2022	3.879	7.327	3.669	2.884
...	...	...	...	...	...
2617	01/08/2013	1.864	7.906	2.030	3.600
2618	01/07/2013	1.901	7.900	2.086	3.590
2619	01/04/2013	1.903	7.930	2.117	3.610
2620	01/03/2013	1.911	7.967	2.073	NaN
2621	01/02/2013	1.835	7.993	1.988	NaN

2622 rows × 5 columns

df_bond['Date'] = pd.to_datetime(df_bond['Date'], infer_datetime_format=True)
df_bond=df_bond.set_index('Date')
df_bond

	US_Price	IN_Price	UK_Price	CHI_Price
Date
2023-01-05	3.722	7.327	3.552	2.870
2023-01-04	3.690	7.321	3.493	2.863
2023-01-03	3.752	7.321	3.653	2.873
2023-01-02	3.879	7.340	3.699	NaN
2022-12-30	3.879	7.327	3.669	2.884
...	...	...	...	...
2013-01-08	1.864	7.906	2.030	3.600
2013-01-07	1.901	7.900	2.086	3.590
2013-01-04	1.903	7.930	2.117	3.610
2013-01-03	1.911	7.967	2.073	NaN
2013-01-02	1.835	7.993	1.988	NaN

2622 rows × 4 columns

df_bond=df_bond.dropna()
df_bond.head()

	US_Price	IN_Price	UK_Price	CHI_Price
Date
2023-01-05	3.722	7.327	3.552	2.870
2023-01-04	3.690	7.321	3.493	2.863
2023-01-03	3.752	7.321	3.653	2.873
2022-12-30	3.879	7.327	3.669	2.884
2022-12-29	3.820	7.311	3.661	2.882

df_bond_return=df_bond.pct_change()
df_bond_return.fillna(0, inplace=True)
df_bond_return.head()

	US_Price	IN_Price	UK_Price	CHI_Price
Date
2023-01-05	0.000000	0.000000	0.000000	0.000000
2023-01-04	-0.008598	-0.000819	-0.016610	-0.002439
2023-01-03	0.016802	0.000000	0.045806	0.003493
2022-12-30	0.033849	0.000820	0.004380	0.003829
2022-12-29	-0.015210	-0.002184	-0.002180	-0.000693

ax1 = plt.subplot(4,2,1)
ax2 = plt.subplot(4,2,2)
ax3 = plt.subplot(4,2,3)
ax4 = plt.subplot(4,2,4)



ax1.plot(df_bond_return['US_Price'], linewidth = 2.5)
ax1.set_ylabel('Return')
ax1.set_title('US Bond Price')

ax2.plot(df_bond_return['IN_Price'], color = 'orange', linewidth = 2.5)
ax2.set_ylabel('Return')
ax2.set_title('India Bond Price')

ax3.plot(df_bond_return['UK_Price'], color = 'green', linewidth = 2.5)
ax3.set_ylabel('Return')
ax3.set_title('United Kingdom Bond Price')



ax4.plot(df_bond_return['CHI_Price'], color = '#DEB887', linewidth = 2.5)
ax4.set_ylabel('Return')
ax4.set_title('China Bond Price')


plt.show()

df_bond_return=df_bond_return.reset_index()
df_bond_return.head()

	Date	US_Price	IN_Price	UK_Price	CHI_Price
0	2023-01-05	0.000000	0.000000	0.000000	0.000000
1	2023-01-04	-0.008598	-0.000819	-0.016610	-0.002439
2	2023-01-03	0.016802	0.000000	0.045806	0.003493
3	2022-12-30	0.033849	0.000820	0.004380	0.003829
4	2022-12-29	-0.015210	-0.002184	-0.002180	-0.000693

def drawdown(return_series: pd.Series):
    """Takes a time series of asset returns.
       returns a DataFrame with columns for
       the wealth index,
       the previous peaks, and
       the percentage drawdown
    """
    wealth_index = 1000*(1+return_series).cumprod()
    previous_peaks = wealth_index.cummax()
    drawdowns = (wealth_index - previous_peaks)/previous_peaks
    return pd.DataFrame({"Wealth": wealth_index,
                         "Previous Peak": previous_peaks,
                         "Drawdown": drawdowns})

FOR UK BOND

filtered_dates1 = ann_vol_portfolio[((ann_vol_portfolio['^FTSE'] < 15) & (ann_vol_portfolio['^FTSE'].diff() <10))]['DateTime']
filtered_dates4  = ann_vol_portfolio[(((27 > ann_vol_portfolio['^FTSE']) & (ann_vol_portfolio['^FTSE'] >= 25)) | ((15 > ann_vol_portfolio['^FTSE'].diff()) & (ann_vol_portfolio['^FTSE'].diff() > 10)))]['DateTime']
filtered_dates3  = ann_vol_portfolio[(((25 > ann_vol_portfolio['^FTSE']) & (ann_vol_portfolio['^FTSE'] >= 20)) | ((30 > ann_vol_portfolio['^FTSE'].diff()) & (ann_vol_portfolio['^FTSE'].diff() > 20)))]['DateTime']
filtered_dates2  = ann_vol_portfolio[(((20 > ann_vol_portfolio['^FTSE']) & (ann_vol_portfolio['^FTSE'] >= 15)) | ((20 > ann_vol_portfolio['^FTSE'].diff()) & (ann_vol_portfolio['^FTSE'].diff() > 15)))]['DateTime']
filtered_dates5  = ann_vol_portfolio[((ann_vol_portfolio['^FTSE']  >=  27) | (ann_vol_portfolio['^FTSE'].diff() >25))]['DateTime']

# Dates fullfilling the condition
def selected_returns_dataframe(filtered_dates):
  selected_returns = []
  for date in filtered_dates:
    return_value = df_bond_return.loc[df_bond_return['Date'] == date, 'UK_Price'].values
    if len(return_value) > 0:
        selected_returns.append((date, return_value[0]))

  selected_return=pd.DataFrame(selected_returns)
  selected_return.set_axis(['Date', 'Portfolio_Return_uk_bond'], axis=1, inplace=True)
  return selected_return

# Define weights
wts_c1 = [0.9, 0.1]
wts_c2 = [0.8, 0.2]
wts_c3 = [0.7, 0.3]
wts_c4 = [0.6, 0.4]
wts_c5 = [0.1, 0.9]
wts_b = [1, 0]

CASE 1

case1_uk=selected_returns_dataframe(filtered_dates1)
case1_uk.head()

	Date	Portfolio_Return_uk_bond
0	2013-02-22	0.014897
1	2013-02-25	0.042585
2	2013-03-01	-0.019916
3	2013-03-04	-0.027523
4	2013-03-06	-0.028798

# Histogram of bond
case1_uk.plot.hist()

<Axes: ylabel='Frequency'>

selected_returns_bond_uk=case1_uk.set_index('Date')
selected_returns_bond_uk.plot()
plt.ylabel('Return')

Text(0, 0.5, 'Return')

case1_ftse=case1_ftse.set_index('Date')
case1_uk=case1_uk.set_index('Date')

case1_data_uk_ftse = case1_ftse.join(case1_uk, how='inner')
case1_data_uk_ftse.head()

	Portfolio_Return_ftse	Portfolio_Return_uk_bond
Date
2013-02-22	0.007025	0.014897
2013-02-25	0.003109	0.042585
2013-03-01	0.002798	-0.019916
2013-03-04	-0.005174	-0.027523
2013-03-06	-0.000684	-0.028798

# UK
weighted_return_c1_uk = (wts_c1 * case1_data_uk_ftse)


#Cumulative Portfolio UK
portfolio_return_c1_uk = weighted_return_c1_uk.sum(axis=1)
cumprod_portfolio_c1_uk = (1 + portfolio_return_c1_uk).cumprod()


#Plotting the solution portfolio's returns
plt.figure(figsize=(10, 5))
plt.plot(cumprod_portfolio_c1_uk, label='CASE 1(90E+10B)')

plt.legend(loc='upper left', fontsize=15)
plt.title("UK Portfolio Performance")
plt.ylabel('Cumulative Return')

Text(0, 0.5, 'Cumulative Return')

CASE 2

case2_uk=selected_returns_dataframe(filtered_dates2)
case2_uk.head()

	Date	Portfolio_Return_uk_bond
0	2013-01-04	0.014861
1	2013-01-07	0.027586
2	2013-01-08	0.002964
3	2013-01-09	-0.034351
4	2013-01-10	0.006241

# Histogram of bond
case2_uk.plot.hist()

<Axes: ylabel='Frequency'>

selected_returns_bond_uk=case2_uk.set_index('Date')
selected_returns_bond_uk.plot()
plt.ylabel('Return')

Text(0, 0.5, 'Return')

case2_ftse=case2_ftse.set_index('Date')
case2_uk=case2_uk.set_index('Date')

case2_data_uk_ftse = case2_ftse.join(case2_uk, how='inner')
case2_data_uk_ftse.head()

	Portfolio_Return_ftse	Portfolio_Return_uk_bond
Date
2013-01-04	0.007028	0.014861
2013-01-07	-0.004138	0.027586
2013-01-08	-0.001814	0.002964
2013-01-09	0.007450	-0.034351
2013-01-10	0.000459	0.006241

# UK

weighted_return_c2_uk = (wts_c2 * case2_data_uk_ftse)


#Cumulative Portfolio UK
portfolio_return_c2_uk = weighted_return_c2_uk.sum(axis=1)
cumprod_portfolio_c2_uk = (1 + portfolio_return_c2_uk).cumprod()

#Plotting the solution portfolio's returns
plt.figure(figsize=(10, 5))
plt.plot(cumprod_portfolio_c2_uk, label='CASE 2(80E+20B)')
plt.legend(loc='upper left', fontsize=15)
plt.title("UK Portfolio Performance C2")
plt.ylabel('Cumulative Return')

Text(0, 0.5, 'Cumulative Return')

CASE 3

case3_uk=selected_returns_dataframe(filtered_dates3)
case3_uk.head()

	Date	Portfolio_Return_uk_bond
0	2016-02-16	-0.028340
1	2016-02-17	0.025606
2	2016-02-18	0.036585
3	2016-02-22	-0.027216
4	2016-02-23	0.051357

# Histogram of bond
case3_uk.plot.hist()

<Axes: ylabel='Frequency'>

selected_returns_bond_uk=case3_uk.set_index('Date')
selected_returns_bond_uk.plot()
plt.ylabel('Return')

Text(0, 0.5, 'Return')

case3_ftse=case3_ftse.set_index('Date')
case3_uk=case3_uk.set_index('Date')

case3_data_uk_ftse = case3_ftse.join(case3_uk, how='inner')
case3_data_uk_ftse.head()

	Portfolio_Return_ftse	Portfolio_Return_uk_bond
Date
2016-02-16	0.027087	-0.028340
2016-02-17	0.028675	0.025606
2016-02-18	-0.009668	0.036585
2016-02-22	0.014705	-0.027216
2016-02-23	-0.012488	0.051357

# UK
weighted_return_c3_uk = (wts_c3 * case3_data_uk_ftse)

#Cumulative Portfolio UK
portfolio_return_c3_uk = weighted_return_c3_uk.sum(axis=1)
cumprod_portfolio_c3_uk = (1 + portfolio_return_c3_uk).cumprod()

#Plotting the solution portfolio's returns
plt.figure(figsize=(10, 5))
plt.plot(cumprod_portfolio_c3_uk, label='CASE 3(70E+30B)')
plt.legend(loc='upper left', fontsize=15)
plt.title("UK Portfolio Performance C3")
plt.ylabel('Cumulative Return')

Text(0, 0.5, 'Cumulative Return')

CASE 4

case4_uk=selected_returns_dataframe(filtered_dates4)
case4_uk.head()

	Date	Portfolio_Return_uk_bond
0	2020-04-07	0.083990
1	2020-04-08	0.249180
2	2020-04-09	-0.001637
3	2020-04-15	0.003333
4	2020-04-16	-0.006623

# Histogram of bond
case4_uk.plot.hist()

<Axes: ylabel='Frequency'>

selected_returns_bond_uk=case4_uk.set_index('Date')
selected_returns_bond_uk.plot()
plt.ylabel('Return')

Text(0, 0.5, 'Return')

case4_ftse=case4_ftse.set_index('Date')
case4_uk=case4_uk.set_index('Date')

case4_data_uk_ftse = case4_ftse.join(case4_uk, how='inner')
case4_data_uk_ftse.head()

	Portfolio_Return_ftse	Portfolio_Return_uk_bond
Date
2020-04-07	0.053365	0.083990
2020-04-08	-0.004698	0.249180
2020-04-09	0.029061	-0.001637
2020-04-15	-0.041933	0.003333
2020-04-16	0.005484	-0.006623

# UK
weighted_return_c4_uk = (wts_c4 * case4_data_uk_ftse)

#Cumulative Portfolio UK
portfolio_return_c4_uk = weighted_return_c4_uk.sum(axis=1)
cumprod_portfolio_c4_uk = (1 + portfolio_return_c4_uk).cumprod()

#Plotting the solution portfolio's returns
plt.figure(figsize=(10, 5))
plt.plot(cumprod_portfolio_c4_uk, label='CASE 4(60E+40B)')
plt.legend(loc='upper left', fontsize=15)
plt.title("UK Portfolio Performance C4")
plt.ylabel('Cumulative Return')

Text(0, 0.5, 'Cumulative Return')

CASE 5

case5_uk=selected_returns_dataframe(filtered_dates5)
case5_uk.head()

	Date	Portfolio_Return_uk_bond
0	2020-05-04	0.126214
1	2020-05-05	-0.112069
2	2020-05-06	-0.002151
3	2020-05-11	0.084337
4	2020-05-12	0.197115

# Histogram of bond
case5_uk.plot.hist()

<Axes: ylabel='Frequency'>

selected_returns_bond_uk=case5_uk.set_index('Date')
selected_returns_bond_uk.plot()
plt.ylabel('Return')

Text(0, 0.5, 'Return')

case5_ftse=case5_ftse.set_index('Date')
case5_uk=case5_uk.set_index('Date')

case5_data_uk_ftse = case5_ftse.join(case5_uk, how='inner')
case5_data_uk_ftse.head()

	Portfolio_Return_ftse	Portfolio_Return_uk_bond
Date
2020-05-04	-0.059114	0.126214
2020-05-05	0.016615	-0.112069
2020-05-06	0.000752	-0.002151
2020-05-11	0.000623	0.084337
2020-05-12	0.009276	0.197115

# UK
weighted_return_c5_uk = (wts_c5 * case5_data_uk_ftse)

#Cumulative Portfolio UK
portfolio_return_c5_uk = weighted_return_c5_uk.sum(axis=1)
cumprod_portfolio_c5_uk = (1 + portfolio_return_c5_uk).cumprod()

#Plotting the solution portfolio's returns
plt.figure(figsize=(10, 5))
plt.plot(cumprod_portfolio_c5_uk, label='CASE 5(10E+90B)')
plt.legend(loc='upper left', fontsize=15)
plt.title("UK Portfolio Performance C5")
plt.ylabel('Cumulative Return')

Text(0, 0.5, 'Cumulative Return')

EQUITY

drawdown(case1_data_uk_ftse['Portfolio_Return_ftse'])

	Wealth	Previous Peak	Drawdown
Date
2013-02-22	1007.025383	1007.025383	0.000000
2013-02-25	1010.156545	1010.156545	0.000000
2013-03-01	1012.983404	1012.983404	0.000000
2013-03-04	1007.742684	1012.983404	-0.005174
2013-03-06	1007.053323	1012.983404	-0.005854
...	...	...	...
2022-02-24	822.254104	1060.320639	-0.224523
2022-02-25	854.484822	1060.320639	-0.194126
2022-02-28	850.925160	1060.320639	-0.197483
2022-03-02	847.650780	1060.320639	-0.200571
2022-03-03	825.893597	1060.320639	-0.221091

1187 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case2_data_uk_ftse['Portfolio_Return_ftse'])

	Wealth	Previous Peak	Drawdown
Date
2013-01-04	1007.027930	1007.027930	0.000000
2013-01-07	1002.860829	1007.027930	-0.004138
2013-01-08	1001.041836	1007.027930	-0.005944
2013-01-09	1008.499726	1008.499726	0.000000
2013-01-10	1008.962710	1008.962710	0.000000
...	...	...	...
2022-12-29	1069.459749	1152.807859	-0.072300
2022-12-30	1060.776181	1152.807859	-0.079833
2023-01-03	1075.353173	1152.807859	-0.067188
2023-01-04	1079.780383	1152.807859	-0.063347
2023-01-05	1086.656033	1152.807859	-0.057383

524 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case3_data_uk_ftse['Portfolio_Return_ftse'])

	Wealth	Previous Peak	Drawdown
Date
2016-02-16	1027.086708	1027.086708	0.000000
2016-02-17	1056.538598	1056.538598	0.000000
2016-02-18	1046.324181	1056.538598	-0.009668
2016-02-22	1061.710784	1061.710784	0.000000
2016-02-23	1048.451860	1061.710784	-0.012488
...	...	...	...
2021-05-28	1111.914387	1252.174457	-0.112013
2021-06-01	1121.081894	1252.174457	-0.104692
2021-06-02	1125.436071	1252.174457	-0.101215
2021-06-03	1118.532705	1252.174457	-0.106728
2021-06-04	1119.261056	1252.174457	-0.106146

189 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case4_data_uk_ftse['Portfolio_Return_ftse'])

	Wealth	Previous Peak	Drawdown
Date
2020-04-07	1053.365340	1053.365340	0.000000
2020-04-08	1048.416618	1053.365340	-0.004698
2020-04-09	1078.884719	1078.884719	0.000000
2020-04-15	1033.644206	1078.884719	-0.041933
2020-04-16	1039.313065	1078.884719	-0.036678
2020-04-17	1068.599391	1078.884719	-0.009533
2020-04-20	1073.363458	1078.884719	-0.005118
2020-04-21	1041.639738	1078.884719	-0.034522
2020-04-22	1065.571064	1078.884719	-0.012340
2020-04-23	1075.911753	1078.884719	-0.002756
2020-04-24	1062.173427	1078.884719	-0.015489
2020-04-27	1079.641733	1079.641733	0.000000
2020-04-28	1100.267750	1100.267750	0.000000
2020-04-29	1129.221643	1129.221643	0.000000
2021-04-19	1126.100795	1129.221643	-0.002764
2021-04-20	1103.546896	1129.221643	-0.022737
2021-04-22	1116.143005	1129.221643	-0.011582
2021-04-23	1116.207337	1129.221643	-0.011525
2021-04-26	1120.148633	1129.221643	-0.008035
2021-04-27	1117.236884	1129.221643	-0.010613
2021-04-28	1120.245170	1129.221643	-0.007949
2021-04-29	1119.891226	1129.221643	-0.008263

Values

Distributions

2-d distributions

Time series

drawdown(case5_data_uk_ftse['Portfolio_Return_ftse'])

	Wealth	Previous Peak	Drawdown
Date
2020-05-04	940.885973	940.885973	0.000000
2020-05-05	956.518910	956.518910	0.000000
2020-05-06	957.238401	957.238401	0.000000
2020-05-11	957.835094	957.835094	0.000000
2020-05-12	966.720448	966.720448	0.000000
...	...	...	...
2021-04-08	1057.871973	1082.225864	-0.022504
2021-04-09	1053.849007	1082.225864	-0.026221
2021-04-12	1049.780431	1082.225864	-0.029980
2021-04-15	1063.949153	1082.225864	-0.016888
2021-04-16	1069.433819	1082.225864	-0.011820

209 rows × 3 columns

Values

Distributions

2-d distributions

Time series

BOND

drawdown(case1_data_uk_ftse['Portfolio_Return_uk_bond'])

	Wealth	Previous Peak	Drawdown
Date
2013-02-22	1014.896684	1014.896684	0.000000
2013-02-25	1058.116232	1058.116232	0.000000
2013-03-01	1037.042639	1058.116232	-0.019916
2013-03-04	1008.500181	1058.116232	-0.046891
2013-03-06	979.456978	1058.116232	-0.074339
...	...	...	...
2022-02-24	5288.909757	9407.017687	-0.437770
2022-02-25	5453.370555	9407.017687	-0.420287
2022-02-28	6104.841247	9407.017687	-0.351033
2022-03-02	5940.353607	9407.017687	-0.368519
2022-03-03	6366.765129	9407.017687	-0.323190

1187 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case2_data_uk_ftse['Portfolio_Return_uk_bond'])

	Wealth	Previous Peak	Drawdown
Date
2013-01-04	1014.860978	1014.860978	0.000000
2013-01-07	1042.857143	1042.857143	0.000000
2013-01-08	1045.948617	1045.948617	0.000000
2013-01-09	1010.019084	1045.948617	-0.034351
2013-01-10	1016.322612	1045.948617	-0.028325
...	...	...	...
2022-12-29	415.062506	1443.217167	-0.712405
2022-12-30	416.880464	1443.217167	-0.711145
2023-01-03	435.976048	1443.217167	-0.697914
2023-01-04	428.734329	1443.217167	-0.702932
2023-01-05	428.734329	1443.217167	-0.702932

524 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case3_data_uk_ftse['Portfolio_Return_uk_bond'])

	Wealth	Previous Peak	Drawdown
Date
2016-02-16	971.659919	971.659919	0.000000
2016-02-17	996.539792	996.539792	0.000000
2016-02-18	1032.998565	1032.998565	0.000000
2016-02-22	1004.884857	1032.998565	-0.027216
2016-02-23	1056.493030	1056.493030	0.000000
...	...	...	...
2021-05-28	653.367945	2249.142809	-0.709504
2021-06-01	675.391584	2249.142809	-0.699712
2021-06-02	642.504345	2249.142809	-0.714334
2021-06-03	682.205118	2249.142809	-0.696682
2021-06-04	667.887233	2249.142809	-0.703048

189 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case4_data_uk_ftse['Portfolio_Return_uk_bond'])

	Wealth	Previous Peak	Drawdown
Date
2020-04-07	1083.989501	1083.989501	0.000000
2020-04-08	1354.098361	1354.098361	0.000000
2020-04-09	1351.882160	1354.098361	-0.001637
2020-04-15	1356.388434	1356.388434	0.000000
2020-04-16	1347.405729	1356.388434	-0.006623
2020-04-17	1211.061102	1356.388434	-0.107143
2020-04-20	1365.491712	1365.491712	0.000000
2020-04-21	1248.210216	1365.491712	-0.085890
2020-04-22	1388.793619	1388.793619	0.000000
2020-04-23	1393.549761	1393.549761	0.000000
2020-04-24	1351.882160	1393.549761	-0.029900
2020-04-27	1422.785071	1422.785071	0.000000
2020-04-28	1427.777299	1427.777299	0.000000
2020-04-29	1792.583834	1792.583834	0.000000
2021-04-19	1851.357074	1851.357074	0.000000
2021-04-20	1828.870956	1851.357074	-0.012146
2021-04-22	1819.051515	1851.357074	-0.017450
2021-04-23	1787.854061	1851.357074	-0.034301
2021-04-26	1746.383220	1851.357074	-0.056701
2021-04-27	1696.111863	1851.357074	-0.083855
2021-04-28	1605.679358	1851.357074	-0.132701
2021-04-29	1607.584079	1851.357074	-0.131673

Values

Distributions

2-d distributions

Time series

drawdown(case5_data_uk_ftse['Portfolio_Return_uk_bond'])

	Wealth	Previous Peak	Drawdown
Date
2020-05-04	1126.213592	1126.213592	0.000000
2020-05-05	1000.000000	1126.213592	-0.112069
2020-05-06	997.849462	1126.213592	-0.113978
2020-05-11	1082.005441	1126.213592	-0.039254
2020-05-12	1295.285360	1295.285360	0.000000
...	...	...	...
2021-04-08	144.992787	3212.307692	-0.954863
2021-04-09	142.787573	3212.307692	-0.955550
2021-04-12	152.655007	3212.307692	-0.952478
2021-04-15	147.074928	3212.307692	-0.954215
2021-04-16	149.020364	3212.307692	-0.953610

209 rows × 3 columns

Values

Distributions

2-d distributions

Time series

a1_uk=case1_data_uk_ftse['Portfolio_Return_ftse'].sum()
a2_uk=case2_data_uk_ftse['Portfolio_Return_ftse'].sum()
a3_uk=case3_data_uk_ftse['Portfolio_Return_ftse'].sum()
a4_uk=case4_data_uk_ftse['Portfolio_Return_ftse'].sum()
a5_uk=case5_data_uk_ftse['Portfolio_Return_ftse'].sum()
a_uk=(a1_uk+a2_uk+a3_uk+a4_uk+a5_uk)

b1_uk=case1_data_uk_ftse['Portfolio_Return_uk_bond'].sum()
b2_uk=case2_data_uk_ftse['Portfolio_Return_uk_bond'].sum()
b3_uk=case3_data_uk_ftse['Portfolio_Return_uk_bond'].sum()
b4_uk=case4_data_uk_ftse['Portfolio_Return_uk_bond'].sum()
b5_uk=case5_data_uk_ftse['Portfolio_Return_uk_bond'].sum()
b_uk=(b1_uk+b2_uk+b3_uk+b4_uk+b5_uk)

c1_uk=filt_case1_ftse['^FTSE'].sum()
c2_uk=filt_case2_ftse['^FTSE'].sum()
c3_uk=filt_case3_ftse['^FTSE'].sum()
c4_uk=filt_case4_ftse['^FTSE'].sum()
c5_uk=filt_case5_ftse['^FTSE'].sum()
c_uk=(c1_uk+c2_uk+c3_uk+c4_uk+c5_uk)

d_uk=return_multiData['^FTSE'].sum()
final_uk=a_uk+b_uk+c_uk-d_uk
final_uk_usd=final_uk*1000000
final_uk_usd_per_change=((final_uk_usd-1000000)/1000000)*100
final_uk_usd_per_change

495.60029262693865

FOR UNITED STATES

filtered_dates1 = ann_vol_portfolio[((ann_vol_portfolio['^GSPC'] < 15) & (ann_vol_portfolio['^GSPC'].diff() <10))]['DateTime']
filtered_dates4  = ann_vol_portfolio[(((27 > ann_vol_portfolio['^GSPC']) & (ann_vol_portfolio['^GSPC'] >= 25)) | ((15 > ann_vol_portfolio['^GSPC'].diff()) & (ann_vol_portfolio['^GSPC'].diff() > 10)))]['DateTime']
filtered_dates3  = ann_vol_portfolio[(((25 > ann_vol_portfolio['^GSPC']) & (ann_vol_portfolio['^GSPC'] >= 20)) | ((30 > ann_vol_portfolio['^GSPC'].diff()) & (ann_vol_portfolio['^GSPC'].diff() > 20)))]['DateTime']
filtered_dates2  = ann_vol_portfolio[(((20 > ann_vol_portfolio['^GSPC']) & (ann_vol_portfolio['^GSPC'] >= 15)) | ((20 > ann_vol_portfolio['^GSPC'].diff()) & (ann_vol_portfolio['^GSPC'].diff() > 15)))]['DateTime']
filtered_dates5  = ann_vol_portfolio[((ann_vol_portfolio['^GSPC']  >=  27) | (ann_vol_portfolio['^GSPC'].diff() >25))]['DateTime']

# Dates fullfilling the condition
def selected_returns_dataframe(filtered_dates):
  selected_returns = []
  for date in filtered_dates:
    return_value = df_bond_return.loc[df_bond_return['Date'] == date, 'US_Price'].values
    if len(return_value) > 0:
        selected_returns.append((date, return_value[0]))

  selected_return=pd.DataFrame(selected_returns)
  selected_return.set_axis(['Date', 'Portfolio_Return_us_bond'], axis=1, inplace=True)
  return selected_return

CASE 1

case1_us=selected_returns_dataframe(filtered_dates1)
case1_us.head()

	Date	Portfolio_Return_us_bond
0	2013-01-23	-0.012959
1	2013-01-24	-0.049769
2	2013-01-25	-0.008142
3	2013-01-29	0.003514
4	2013-01-30	0.003526

# Histogram of bond
case1_us.plot.hist()

<Axes: ylabel='Frequency'>

selected_returns_bond_us=case1_us.set_index('Date')
selected_returns_bond_us.plot()
plt.ylabel('Return')

Text(0, 0.5, 'Return')

case1_gspc=case1_gspc.set_index('Date')
case1_us=case1_us.set_index('Date')

case1_data_us_gspc = case1_gspc.join(case1_us, how='inner')
case1_data_us_gspc.head()

	Portfolio_Return_gspc	Portfolio_Return_us_bond
Date
2013-01-23	0.001507	-0.012959
2013-01-24	0.000007	-0.049769
2013-01-25	0.005445	-0.008142
2013-01-29	0.003247	0.003514
2013-01-30	-0.003900	0.003526

# US
weighted_return_c1_us = (wts_c1 * case1_data_us_gspc)

#Cumulative Portfolio US
portfolio_return_c1_us = weighted_return_c1_us.sum(axis=1)
cumprod_portfolio_c1_us = (1 + portfolio_return_c1_us).cumprod()

#Plotting the solution portfolio's returns
plt.figure(figsize=(10, 5))
plt.plot(cumprod_portfolio_c1_us, label='CASE 1(90E+10B)')
plt.legend(loc='upper left', fontsize=15)
plt.title("US Portfolio Performance C1")
plt.ylabel('Cumulative Return')

Text(0, 0.5, 'Cumulative Return')

CASE 2

case2_us=selected_returns_dataframe(filtered_dates2)
case2_us.head()

	Date	Portfolio_Return_us_bond
0	2013-01-04	0.001052
1	2013-01-07	0.019850
2	2013-01-08	0.002151
3	2013-01-09	-0.020537
4	2013-01-10	0.017685

# Histogram of bond
case2_us.plot.hist()

<Axes: ylabel='Frequency'>

selected_returns_bond_us=case2_us.set_index('Date')
selected_returns_bond_us.plot()
plt.ylabel('Return')

Text(0, 0.5, 'Return')

case2_gspc=case2_gspc.set_index('Date')
case2_us=case2_us.set_index('Date')

case2_data_us_gspc = case2_gspc.join(case2_us, how='inner')
case2_data_us_gspc.head()

	Portfolio_Return_gspc	Portfolio_Return_us_bond
Date
2013-01-04	0.004865	0.001052
2013-01-07	-0.003123	0.019850
2013-01-08	-0.003242	0.002151
2013-01-09	0.002656	-0.020537
2013-01-10	0.007597	0.017685

# US
weighted_return_c2_us = (wts_c2 * case2_data_us_gspc)

#Cumulative Portfolio US

portfolio_return_c2_us = weighted_return_c2_us.sum(axis=1)
cumprod_portfolio_c2_us = (1 + portfolio_return_c2_us).cumprod()

#Plotting the solution portfolio's returns
plt.figure(figsize=(10, 5))
plt.plot(cumprod_portfolio_c2_us, label='CASE 2(80E+20B)')
plt.legend(loc='upper left', fontsize=15)
plt.title("US Portfolio Performance C2")
plt.ylabel('Cumulative Return')

Text(0, 0.5, 'Cumulative Return')

CASE 3

case3_us=selected_returns_dataframe(filtered_dates3)
case3_us.head()

	Date	Portfolio_Return_us_bond
0	2020-03-12	-0.173957
1	2020-03-13	0.328378
2	2021-04-28	-0.018282
3	2021-04-29	0.009225
4	2021-04-30	0.035669

# Histogram of bond
case3_us.plot.hist()

<Axes: ylabel='Frequency'>

selected_returns_bond_us=case3_us.set_index('Date')
selected_returns_bond_us.plot()
plt.ylabel('Return')

Text(0, 0.5, 'Return')

case3_gspc=case3_gspc.set_index('Date')
case3_us=case3_us.set_index('Date')

case3_data_us_gspc = case3_gspc.join(case3_us, how='inner')
case3_data_us_gspc.head()

	Portfolio_Return_gspc	Portfolio_Return_us_bond
Date
2020-03-12	-0.095113	-0.173957
2020-03-13	0.092871	0.328378
2021-04-28	-0.000846	-0.018282
2021-04-29	0.006763	0.009225
2021-04-30	-0.007195	0.035669

# US
weighted_return_c3_us = (wts_c3 * case3_data_us_gspc)


#Cumulative Portfolio US
portfolio_return_c3_us = weighted_return_c3_us.sum(axis=1)
cumprod_portfolio_c3_us = (1 + portfolio_return_c3_us).cumprod()

#Plotting the solution portfolio's returns
plt.figure(figsize=(10, 5))
plt.plot(cumprod_portfolio_c3_us, label='CASE 3(70E+30B)')
plt.legend(loc='upper left', fontsize=15)
plt.title("US Portfolio Performance C3")
plt.ylabel('Cumulative Return')

Text(0, 0.5, 'Cumulative Return')

CASE 4

case4_us=selected_returns_dataframe(filtered_dates4)
case4_us.head()

	Date	Portfolio_Return_us_bond
0	2020-03-16	-0.307764
1	2020-03-17	-0.096365
2	2020-03-18	0.021589
3	2020-03-19	0.308475
4	2021-04-22	-0.010904

# Histogram of bond
case4_us.plot.hist()

<Axes: ylabel='Frequency'>

selected_returns_bond_us=case4_us.set_index('Date')
selected_returns_bond_us.plot()
plt.ylabel('Return')

Text(0, 0.5, 'Return')

case4_gspc=case4_gspc.set_index('Date')
case4_us=case4_us.set_index('Date')

case4_data_us_gspc = case4_gspc.join(case4_us, how='inner')
case4_data_us_gspc.head()

	Portfolio_Return_gspc	Portfolio_Return_us_bond
Date
2020-03-16	-0.119841	-0.307764
2020-03-17	0.059955	-0.096365
2020-03-18	-0.051831	0.021589
2020-03-19	0.004708	0.308475
2021-04-22	0.000010	-0.010904

# US
weighted_return_c4_us = (wts_c4 * case4_data_us_gspc)

#Cumulative Portfolio US

portfolio_return_c4_us = weighted_return_c4_us.sum(axis=1)
cumprod_portfolio_c4_us = (1 + portfolio_return_c4_us).cumprod()

#Plotting the solution portfolio's returns
plt.figure(figsize=(10, 5))
plt.plot(cumprod_portfolio_c4_us, label='CASE 4(60E+40B)')
plt.legend(loc='upper left', fontsize=15)
plt.title("US Portfolio Performance C4")
plt.ylabel('Cumulative Return')

Text(0, 0.5, 'Cumulative Return')

CASE 5

case5_us=selected_returns_dataframe(filtered_dates5)
case5_us.head()

	Date	Portfolio_Return_us_bond
0	2020-03-20	0.159895
1	2020-03-23	-0.110723
2	2020-03-24	0.011792
3	2020-03-26	0.254438
4	2020-03-27	-0.068871

# Histogram of bond
case5_us.plot.hist()

<Axes: ylabel='Frequency'>

selected_returns_bond_us=case5_us.set_index('Date')
selected_returns_bond_us.plot()
plt.ylabel('Return')

Text(0, 0.5, 'Return')

case5_gspc=case5_gspc.set_index('Date')
case5_us=case5_us.set_index('Date')

case5_data_us_gspc = case5_gspc.join(case5_us, how='inner')
case5_data_us_gspc.head()

	Portfolio_Return_gspc	Portfolio_Return_us_bond
Date
2020-03-20	-0.043360	0.159895
2020-03-23	-0.029294	-0.110723
2020-03-24	0.093828	0.011792
2020-03-26	0.062414	0.254438
2020-03-27	-0.033687	-0.068871

# US
weighted_return_c5_us = (wts_c5 * case5_data_us_gspc)

#Cumulative Portfolio US
portfolio_return_c5_us = weighted_return_c5_us.sum(axis=1)
cumprod_portfolio_c5_us = (1 + portfolio_return_c5_us).cumprod()

#Plotting the solution portfolio's returns
plt.figure(figsize=(10, 5))
plt.plot(cumprod_portfolio_c5_us, label='CASE 5(10E+90B)')
plt.legend(loc='upper left', fontsize=15)
plt.title("US Portfolio Performance C5")
plt.ylabel('Cumulative Return')

Text(0, 0.5, 'Cumulative Return')

EQUITY

drawdown(case1_data_us_gspc['Portfolio_Return_gspc'])

	Wealth	Previous Peak	Drawdown
Date
2013-01-23	1001.507477	1001.507477	0.000000
2013-01-24	1001.514101	1001.514101	0.000000
2013-01-25	1006.967828	1006.967828	0.000000
2013-01-29	1010.237382	1010.237382	0.000000
2013-01-30	1006.297838	1010.237382	-0.003900
...	...	...	...
2022-01-24	1320.332028	1761.700133	-0.250535
2022-01-25	1304.261070	1761.700133	-0.259658
2022-01-27	1295.297319	1761.700133	-0.264746
2022-01-28	1326.834760	1761.700133	-0.246844
2022-02-08	1337.981774	1761.700133	-0.240517

1042 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case2_data_us_gspc['Portfolio_Return_gspc'])

	Wealth	Previous Peak	Drawdown
Date
2013-01-04	1004.865097	1004.865097	0.000000
2013-01-07	1001.726786	1004.865097	-0.003123
2013-01-08	998.478816	1004.865097	-0.006355
2013-01-09	1001.130642	1004.865097	-0.003716
2013-01-10	1008.736647	1008.736647	0.000000
...	...	...	...
2022-06-08	1328.581544	1567.695656	-0.152526
2022-06-09	1296.963039	1567.695656	-0.172695
2022-06-10	1259.208051	1567.695656	-0.196778
2022-06-14	1205.822905	1567.695656	-0.230831
2022-06-15	1223.418881	1567.695656	-0.219607

716 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case3_data_us_gspc['Portfolio_Return_gspc'])

	Wealth	Previous Peak	Drawdown
Date
2020-03-12	904.887319	904.887319	0.000000
2020-03-13	988.925335	988.925335	0.000000
2021-04-28	988.089159	988.925335	-0.000846
2021-04-29	994.771415	994.771415	0.000000
2021-04-30	987.614327	994.771415	-0.007195
...	...	...	...
2022-12-29	962.138581	1087.362694	-0.115163
2022-12-30	959.694035	1087.362694	-0.117411
2023-01-03	955.854732	1087.362694	-0.120942
2023-01-04	963.060893	1087.362694	-0.114315
2023-01-05	951.845539	1087.362694	-0.124629

132 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case4_data_us_gspc['Portfolio_Return_gspc'])

	Wealth	Previous Peak	Drawdown
Date
2020-03-16	880.159448	880.159448	0.000000
2020-03-17	932.929275	932.929275	0.000000
2020-03-18	884.574839	932.929275	-0.051831
2020-03-19	888.739248	932.929275	-0.047367
2021-04-22	888.747853	932.929275	-0.047358
2021-04-23	898.460709	932.929275	-0.036947
2021-04-26	900.062009	932.929275	-0.035230
2021-04-27	899.868590	932.929275	-0.035438

Values

Distributions

2-d distributions

Time series

drawdown(case5_data_us_gspc['Portfolio_Return_gspc'])

	Wealth	Previous Peak	Drawdown
Date
2020-03-20	956.640488	956.640488	0.000000
2020-03-23	928.616787	956.640488	-0.029294
2020-03-24	1015.746802	1015.746802	0.000000
2020-03-26	1079.143788	1079.143788	0.000000
2020-03-27	1042.790291	1079.143788	-0.033687
...	...	...	...
2021-04-12	1727.184598	1727.523329	-0.000196
2021-04-15	1739.207726	1739.207726	0.000000
2021-04-16	1745.484213	1745.484213	0.000000
2021-04-19	1736.221696	1745.484213	-0.005307
2021-04-20	1724.411361	1745.484213	-0.012073

233 rows × 3 columns

Values

Distributions

2-d distributions

Time series

BOND

drawdown(case1_data_us_gspc['Portfolio_Return_us_bond'])

	Wealth	Previous Peak	Drawdown
Date
2013-01-23	987.041037	987.041037	0.000000
2013-01-24	937.916880	987.041037	-0.049769
2013-01-25	930.279898	987.041037	-0.057506
2013-01-29	933.548954	987.041037	-0.054194
2013-01-30	936.841066	987.041037	-0.050859
...	...	...	...
2022-01-24	959.248077	1333.931642	-0.280887
2022-01-25	944.883297	1333.931642	-0.291655
2022-01-27	958.169058	1333.931642	-0.281696
2022-01-28	866.984521	1333.931642	-0.350053
2022-02-08	875.899529	1333.931642	-0.343370

1042 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case2_data_us_gspc['Portfolio_Return_us_bond'])

	Wealth	Previous Peak	Drawdown
Date
2013-01-04	1001.052078	1001.052078	0.000000
2013-01-07	1020.922747	1020.922747	0.000000
2013-01-08	1023.118280	1023.118280	0.000000
2013-01-09	1002.106372	1023.118280	-0.020537
2013-01-10	1019.828510	1023.118280	-0.003215
...	...	...	...
2022-06-08	1534.363114	2971.796373	-0.483692
2022-06-09	1476.673000	2971.796373	-0.503104
2022-06-10	1390.145760	2971.796373	-0.532220
2022-06-14	1469.112120	2971.796373	-0.505648
2022-06-15	1513.714272	2971.796373	-0.490640

716 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case3_data_us_gspc['Portfolio_Return_us_bond'])

	Wealth	Previous Peak	Drawdown
Date
2020-03-12	826.042726	826.042726	0.000000
2020-03-13	1097.297297	1097.297297	0.000000
2021-04-28	1077.237018	1097.297297	-0.018282
2021-04-29	1087.174629	1097.297297	-0.009225
2021-04-30	1125.952832	1125.952832	0.000000
...	...	...	...
2022-12-29	959.614632	1302.768070	-0.263403
2022-12-30	992.096257	1302.768070	-0.238471
2023-01-03	1008.765625	1302.768070	-0.225675
2023-01-04	1000.092734	1302.768070	-0.232332
2023-01-05	1000.092734	1302.768070	-0.232332

132 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case4_data_us_gspc['Portfolio_Return_us_bond'])

	Wealth	Previous Peak	Drawdown
Date
2020-03-16	692.235734	692.235734	0.000000
2020-03-17	625.528318	692.235734	-0.096365
2020-03-18	639.032815	692.235734	-0.076857
2020-03-19	836.158192	836.158192	0.000000
2021-04-22	827.040367	836.158192	-0.010904
2021-04-23	822.293324	836.158192	-0.016582
2021-04-26	793.449804	836.158192	-0.051077
2021-04-27	800.345085	836.158192	-0.042831

Values

Distributions

2-d distributions

Time series

drawdown(case5_data_us_gspc['Portfolio_Return_us_bond'])

	Wealth	Previous Peak	Drawdown
Date
2020-03-20	1159.895151	1159.895151	0.000000
2020-03-23	1031.468531	1159.895151	-0.110723
2020-03-24	1043.632075	1159.895151	-0.100236
2020-03-26	1309.171598	1309.171598	0.000000
2020-03-27	1219.008264	1309.171598	-0.068871
...	...	...	...
2021-04-12	376.151426	1477.462437	-0.745407
2021-04-15	372.366255	1477.462437	-0.747969
2021-04-16	367.740587	1477.462437	-0.751100
2021-04-19	378.073017	1477.462437	-0.744107
2021-04-20	383.957422	1477.462437	-0.740124

233 rows × 3 columns

Values

Distributions

2-d distributions

Time series

a1_us=case1_data_us_gspc['Portfolio_Return_gspc'].sum()
a2_us=case2_data_us_gspc['Portfolio_Return_gspc'].sum()
a3_us=case3_data_us_gspc['Portfolio_Return_gspc'].sum()
a4_us=case4_data_us_gspc['Portfolio_Return_gspc'].sum()
a5_us=case5_data_us_gspc['Portfolio_Return_gspc'].sum()
a_us=(a1_us+a2_us+a3_us+a4_us+a5_us)

b1_us=case1_data_us_gspc['Portfolio_Return_us_bond'].sum()
b2_us=case2_data_us_gspc['Portfolio_Return_us_bond'].sum()
b3_us=case3_data_us_gspc['Portfolio_Return_us_bond'].sum()
b4_us=case4_data_us_gspc['Portfolio_Return_us_bond'].sum()
b5_us=case5_data_us_gspc['Portfolio_Return_us_bond'].sum()
b_us=(b1_us+b2_us+b3_us+b4_us+b5_us)

c1_us=filt_case1_gspc['^GSPC'].sum()
c2_us=filt_case2_gspc['^GSPC'].sum()
c3_us=filt_case3_gspc['^GSPC'].sum()
c4_us=filt_case4_gspc['^GSPC'].sum()
c5_us=filt_case5_gspc['^GSPC'].sum()
c_us=(c1_us+c2_us+c3_us+c4_us+c5_us)

d_us=return_multiData['^GSPC'].sum()
final_us=a_us+b_us+c_us-d_us
final_us_usd=final_us*1000000
final_us_usd_per_change=((final_us_usd-1000000)/1000000)*100
final_us_usd_per_change

570.4119952300248

FOR INDIA

filtered_dates1 = ann_vol_portfolio[((ann_vol_portfolio['^NSEI'] < 15) & (ann_vol_portfolio['^NSEI'].diff() <10))]['DateTime']
filtered_dates4  = ann_vol_portfolio[(((27 > ann_vol_portfolio['^NSEI']) & (ann_vol_portfolio['^NSEI'] >= 25)) | ((15 > ann_vol_portfolio['^NSEI'].diff()) & (ann_vol_portfolio['^NSEI'].diff() > 10)))]['DateTime']
filtered_dates3  = ann_vol_portfolio[(((25 > ann_vol_portfolio['^NSEI']) & (ann_vol_portfolio['^NSEI'] >= 20)) | ((30 > ann_vol_portfolio['^NSEI'].diff()) & (ann_vol_portfolio['^NSEI'].diff() > 20)))]['DateTime']
filtered_dates2  = ann_vol_portfolio[(((20 > ann_vol_portfolio['^NSEI']) & (ann_vol_portfolio['^NSEI'] >= 15)) | ((20 > ann_vol_portfolio['^NSEI'].diff()) & (ann_vol_portfolio['^NSEI'].diff() > 15)))]['DateTime']
filtered_dates5  = ann_vol_portfolio[((ann_vol_portfolio['^NSEI']  >=  27) | (ann_vol_portfolio['^NSEI'].diff() >25))]['DateTime']

# Dates fullfilling the condition
def selected_returns_dataframe(filtered_dates):
  selected_returns = []
  for date in filtered_dates:
    return_value = df_bond_return.loc[df_bond_return['Date'] == date, 'IN_Price'].values
    if len(return_value) > 0:
        selected_returns.append((date, return_value[0]))

  selected_return=pd.DataFrame(selected_returns)
  selected_return.set_axis(['Date', 'Portfolio_Return_IN_bond'], axis=1, inplace=True)
  return selected_return

CASE 1

case1_in=selected_returns_dataframe(filtered_dates1)
case1_in.head()

	Date	Portfolio_Return_IN_bond
0	2014-11-11	0.003310
1	2014-11-12	-0.007663
2	2014-11-13	0.000730
3	2014-11-14	0.004156
4	2014-11-17	0.003434

# Histogram of bond
case1_in.plot.hist()

<Axes: ylabel='Frequency'>

selected_returns_bond_in=case1_in.set_index('Date')
selected_returns_bond_in.plot()
plt.ylabel('Return')

Text(0, 0.5, 'Return')

case1_nsei=case1_nsei.set_index('Date')
case1_in=case1_in.set_index('Date')

case1_data_in_nsei = case1_nsei.join(case1_in, how='inner')
case1_data_in_nsei.head()

	Portfolio_Return_nsei	Portfolio_Return_IN_bond
Date
2014-11-11	0.002205	0.003310
2014-11-12	0.002469	-0.007663
2014-11-13	-0.003036	0.000730
2014-11-14	0.003835	0.004156
2014-11-17	0.004869	0.003434

# India
weighted_return_c1_in = (wts_c1 * case1_data_in_nsei)

#Cumulative Portfolio in
portfolio_return_c1_in = weighted_return_c1_in.sum(axis=1)
cumprod_portfolio_c1_in = (1 + portfolio_return_c1_in).cumprod()

#Plotting the solution portfolio's returns
plt.figure(figsize=(10, 5))
plt.plot(cumprod_portfolio_c1_in, label='CASE 1(90E+10B)')
plt.legend(loc='upper left', fontsize=15)
plt.title("India Portfolio Performance C1")
plt.ylabel('Cumulative Return')

Text(0, 0.5, 'Cumulative Return')

CASE 2

case2_in=selected_returns_dataframe(filtered_dates2)
case2_in.head()

	Date	Portfolio_Return_IN_bond
0	2013-01-04	0.003797
1	2013-01-07	-0.000759
2	2013-01-08	0.001140
3	2013-01-09	0.002539
4	2013-01-10	0.001398

# Histogram of bond
case2_in.plot.hist()

<Axes: ylabel='Frequency'>

selected_returns_bond_in=case2_in.set_index('Date')
selected_returns_bond_in.plot()
plt.ylabel('Return')

Text(0, 0.5, 'Return')

case2_nsei=case2_nsei.set_index('Date')
case2_in=case2_in.set_index('Date')

case2_data_in_nsei = case2_nsei.join(case2_in, how='inner')
case2_data_in_nsei.head()

	Portfolio_Return_nsei	Portfolio_Return_IN_bond
Date
2013-01-04	0.001107	0.003797
2013-01-07	-0.004613	-0.000759
2013-01-08	0.002221	0.001140
2013-01-09	-0.005032	0.002539
2013-01-10	-0.000477	0.001398

# India
weighted_return_c2_in = (wts_c2 * case2_data_in_nsei)

#Cumulative Portfolio india
portfolio_return_c2_in = weighted_return_c2_in.sum(axis=1)
cumprod_portfolio_c2_in = (1 + portfolio_return_c2_in).cumprod()

#Plotting the solution portfolio's returns
plt.figure(figsize=(10, 5))
plt.plot(cumprod_portfolio_c2_in, label='CASE 2(80E+20B)')
plt.legend(loc='upper left', fontsize=15)
plt.title("India Portfolio Performance C2")
plt.ylabel('Cumulative Return')

Text(0, 0.5, 'Cumulative Return')

CASE 3

case3_in=selected_returns_dataframe(filtered_dates3)
case3_in.head()

	Date	Portfolio_Return_IN_bond
0	2020-03-16	-0.009098
1	2020-03-17	-0.004924
2	2020-03-18	-0.017785
3	2020-03-19	0.024125
4	2020-03-20	-0.018812

# Histogram of bond
case3_in.plot.hist()

<Axes: ylabel='Frequency'>

selected_returns_bond_in=case3_in.set_index('Date')
selected_returns_bond_in.plot()
plt.ylabel('Return')

Text(0, 0.5, 'Return')

case3_nsei=case3_nsei.set_index('Date')
case3_in=case3_in.set_index('Date')

case3_data_in_nsei = case3_nsei.join(case3_in, how='inner')
case3_data_in_nsei.head()

	Portfolio_Return_nsei	Portfolio_Return_IN_bond
Date
2020-03-16	-0.076121	-0.009098
2020-03-17	-0.025045	-0.004924
2020-03-18	-0.055565	-0.017785
2020-03-19	-0.024248	0.024125
2020-03-20	0.058329	-0.018812

# India
weighted_return_c3_in = (wts_c3 * case3_data_in_nsei)

#Cumulative Portfolio india

portfolio_return_c3_in = weighted_return_c3_in.sum(axis=1)
cumprod_portfolio_c3_in = (1 + portfolio_return_c3_in).cumprod()

#Plotting the solution portfolio's returns
plt.figure(figsize=(10, 5))
plt.plot(cumprod_portfolio_c3_in, label='CASE 3(70E+30B)')
plt.legend(loc='upper left', fontsize=15)
plt.title("India Portfolio Performance C3")
plt.ylabel('Cumulative Return')

Text(0, 0.5, 'Cumulative Return')

CASE 4

case4_in=selected_returns_dataframe(filtered_dates4)
case4_in.head()

	Date	Portfolio_Return_IN_bond
0	2020-03-23	0.011897
1	2020-03-24	0.012691

# Histogram of bond
case4_in.plot.hist()

<Axes: ylabel='Frequency'>

selected_returns_bond_in=case4_in.set_index('Date')
selected_returns_bond_in.plot()
plt.ylabel('Return')

Text(0, 0.5, 'Return')

case4_nsei=case4_nsei.set_index('Date')
case4_in=case4_in.set_index('Date')

case4_data_in_nsei = case4_nsei.join(case4_in, how='inner')
case4_data_in_nsei.head()

	Portfolio_Return_nsei	Portfolio_Return_IN_bond
Date
2020-03-23	-0.129805	0.011897
2020-03-24	0.025071	0.012691

# India
weighted_return_c4_in = (wts_c4 * case4_data_in_nsei)

#Cumulative Portfolio india
portfolio_return_c4_in = weighted_return_c4_in.sum(axis=1)
cumprod_portfolio_c4_in = (1 + portfolio_return_c4_in).cumprod()

#Plotting the solution portfolio's returns
plt.figure(figsize=(10, 5))
plt.plot(cumprod_portfolio_c4_in, label='CASE 4(60E+40B)')
plt.legend(loc='upper left', fontsize=15)
plt.title("India Portfolio Performance C4")
plt.ylabel('Cumulative Return')

Text(0, 0.5, 'Cumulative Return')

CASE 5

case5_in=selected_returns_dataframe(filtered_dates5)
case5_in.head()

	Date	Portfolio_Return_IN_bond
0	2020-03-26	0.013184
1	2020-03-27	-0.010469
2	2020-03-30	0.011567
3	2020-03-31	-0.026641
4	2020-04-03	-0.016838

# Histogram of bond
case5_in.plot.hist()

<Axes: ylabel='Frequency'>

selected_returns_bond_in=case5_in.set_index('Date')
selected_returns_bond_in.plot()
plt.ylabel('Return')

Text(0, 0.5, 'Return')

case5_nsei=case5_nsei.set_index('Date')
case5_in=case5_in.set_index('Date')

case5_data_in_nsei = case5_nsei.join(case5_in, how='inner')
case5_data_in_nsei.head()

	Portfolio_Return_nsei	Portfolio_Return_IN_bond
Date
2020-03-26	0.038904	0.013184
2020-03-27	0.002176	-0.010469
2020-03-30	-0.043781	0.011567
2020-03-31	0.038238	-0.026641
2020-04-03	-0.020597	-0.016838

# India
weighted_return_c5_in = (wts_c5 * case5_data_in_nsei)

#Cumulative Portfolio india

portfolio_return_c5_in = weighted_return_c5_in.sum(axis=1)
cumprod_portfolio_c5_in = (1 + portfolio_return_c5_in).cumprod()

#Plotting the solution portfolio's returns
plt.figure(figsize=(10, 5))
plt.plot(cumprod_portfolio_c5_in, label='CASE 5(10E+90B)')
plt.legend(loc='upper left', fontsize=15)
plt.title("India Portfolio Performance C5")
plt.ylabel('Cumulative Return')

Text(0, 0.5, 'Cumulative Return')

EQUITY

drawdown(case1_data_in_nsei['Portfolio_Return_nsei'])

	Wealth	Previous Peak	Drawdown
Date
2014-11-11	1002.205158	1002.205158	0.000000
2014-11-12	1004.679846	1004.679846	0.000000
2014-11-13	1001.629818	1004.679846	-0.003036
2014-11-14	1005.470880	1005.470880	0.000000
2014-11-17	1010.366420	1010.366420	0.000000
...	...	...	...
2020-02-03	1242.113700	1420.829473	-0.125783
2020-02-13	1239.410861	1420.829473	-0.127685
2020-02-14	1233.180523	1420.829473	-0.132070
2020-02-18	1220.867481	1420.829473	-0.140736
2020-02-20	1216.331650	1420.829473	-0.143928

626 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case2_data_in_nsei['Portfolio_Return_nsei'])

	Wealth	Previous Peak	Drawdown
Date
2013-01-04	1001.106565	1001.106565	0.000000
2013-01-07	996.488876	1001.106565	-0.004613
2013-01-08	998.702088	1001.106565	-0.002402
2013-01-09	993.676679	1001.106565	-0.007422
2013-01-10	993.202413	1001.106565	-0.007895
...	...	...	...
2022-12-29	1603.012433	1788.724077	-0.103824
2022-12-30	1595.460517	1788.724077	-0.108045
2023-01-03	1606.673937	1788.724077	-0.101777
2023-01-04	1589.966023	1788.724077	-0.111117
2023-01-05	1585.489571	1788.724077	-0.113620

1249 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case3_data_in_nsei['Portfolio_Return_nsei'])

	Wealth	Previous Peak	Drawdown
Date
2020-03-16	923.878999	923.878999	0.000000
2020-03-17	900.740279	923.878999	-0.025045
2020-03-18	850.691060	923.878999	-0.079218
2020-03-19	830.063689	923.878999	-0.101545
2020-03-20	878.480595	923.878999	-0.049139
2021-04-29	880.274286	923.878999	-0.047197
2021-04-30	864.683913	923.878999	-0.064072
2021-05-06	871.010296	923.878999	-0.057225
2021-05-07	876.827990	923.878999	-0.050928
2021-05-10	883.878934	923.878999	-0.043296
2021-05-11	878.460579	923.878999	-0.049161
2021-05-12	869.336289	923.878999	-0.059037
2021-05-14	868.230124	923.878999	-0.060234
2021-05-17	882.743251	923.878999	-0.044525
2021-05-18	893.683479	923.878999	-0.032683
2021-05-20	881.731706	923.878999	-0.045620
2021-05-21	897.658546	923.878999	-0.028381
2021-05-24	898.983587	923.878999	-0.026947
2021-05-25	899.619478	923.878999	-0.026258

Values

Distributions

2-d distributions

Time series

drawdown(case4_data_in_nsei['Portfolio_Return_nsei'])

	Wealth	Previous Peak	Drawdown
Date
2020-03-23	870.195339	870.195339	0.0
2020-03-24	892.012376	892.012376	0.0

Values

Distributions

2-d distributions

Time series

drawdown(case5_data_in_nsei['Portfolio_Return_nsei'])

	Wealth	Previous Peak	Drawdown
Date
2020-03-26	1038.904356	1038.904356	0.000000
2020-03-27	1041.164533	1041.164533	0.000000
2020-03-30	995.581791	1041.164533	-0.043781
2020-03-31	1033.650571	1041.164533	-0.007217
2020-04-03	1012.360911	1041.164533	-0.027665
...	...	...	...
2021-04-22	1680.092400	1823.763937	-0.078777
2021-04-23	1672.535121	1823.763937	-0.082921
2021-04-26	1689.288099	1823.763937	-0.073735
2021-04-27	1708.886618	1823.763937	-0.062989
2021-04-28	1733.552440	1823.763937	-0.049464

235 rows × 3 columns

Values

Distributions

2-d distributions

Time series

BOND

drawdown(case1_data_in_nsei['Portfolio_Return_IN_bond'])

	Wealth	Previous Peak	Drawdown
Date
2014-11-11	1003.309635	1003.309635	0.000000
2014-11-12	995.620971	1003.309635	-0.007663
2014-11-13	996.348144	1003.309635	-0.006939
2014-11-14	1000.488938	1003.309635	-0.002811
2014-11-17	1003.924936	1003.924936	0.000000
...	...	...	...
2020-02-03	1114.229826	1138.180174	-0.021043
2020-02-13	1123.676899	1138.180174	-0.012743
2020-02-14	1120.510125	1138.180174	-0.015525
2020-02-18	1113.879888	1138.180174	-0.021350
2020-02-20	1123.501907	1138.180174	-0.012896

626 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case2_data_in_nsei['Portfolio_Return_IN_bond'])

	Wealth	Previous Peak	Drawdown
Date
2013-01-04	1003.797468	1003.797468	0.000000
2013-01-07	1003.035669	1003.797468	-0.000759
2013-01-08	1004.178802	1004.178802	0.000000
2013-01-09	1006.728450	1006.728450	0.000000
2013-01-10	1008.136283	1008.136283	0.000000
...	...	...	...
2022-12-29	1086.717522	1361.652366	-0.201913
2022-12-30	1087.608152	1361.652366	-0.201259
2023-01-03	1087.608152	1361.652366	-0.201259
2023-01-04	1086.717522	1361.652366	-0.201913
2023-01-05	1086.717522	1361.652366	-0.201913

1249 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case3_data_in_nsei['Portfolio_Return_IN_bond'])

	Wealth	Previous Peak	Drawdown
Date
2020-03-16	990.901836	990.901836	0.000000
2020-03-17	986.022872	990.901836	-0.004924
2020-03-18	968.486739	990.901836	-0.022621
2020-03-19	991.851734	991.851734	0.000000
2020-03-20	973.193290	991.851734	-0.018812
2021-04-29	977.873656	991.851734	-0.014093
2021-04-30	987.040199	991.851734	-0.004851
2021-05-06	980.312244	991.851734	-0.011634
2021-05-07	980.475249	991.851734	-0.011470
2021-05-10	981.454419	991.851734	-0.010483
2021-05-11	981.127811	991.851734	-0.010812
2021-05-12	985.061501	991.851734	-0.006846
2021-05-14	987.370755	991.851734	-0.004518
2021-05-17	986.709864	991.851734	-0.005184
2021-05-18	986.875004	991.851734	-0.005018
2021-05-20	985.388995	991.851734	-0.006516
2021-05-21	986.214002	991.851734	-0.005684
2021-05-24	986.214002	991.851734	-0.005684
2021-05-25	983.743114	991.851734	-0.008175

Values

Distributions

2-d distributions

Time series

drawdown(case4_data_in_nsei['Portfolio_Return_IN_bond'])

	Wealth	Previous Peak	Drawdown
Date
2020-03-23	1011.897208	1011.897208	0.0
2020-03-24	1024.738956	1024.738956	0.0

Values

Distributions

2-d distributions

Time series

drawdown(case5_data_in_nsei['Portfolio_Return_IN_bond'])

	Wealth	Previous Peak	Drawdown
Date
2020-03-26	1013.183594	1013.183594	0.000000
2020-03-27	1002.576904	1013.183594	-0.010469
2020-03-30	1014.173998	1014.173998	0.000000
2020-03-31	987.155090	1014.173998	-0.026641
2020-04-03	970.533209	1014.173998	-0.043031
...	...	...	...
2021-04-22	1030.554068	1094.712388	-0.058607
2021-04-23	1030.042034	1094.712388	-0.059075
2021-04-26	1027.150091	1094.712388	-0.061717
2021-04-27	1027.489476	1094.712388	-0.061407
2021-04-28	1026.471992	1094.712388	-0.062336

235 rows × 3 columns

Values

Distributions

2-d distributions

Time series

a1_in=case1_data_in_nsei['Portfolio_Return_nsei'].sum()
a2_in=case2_data_in_nsei['Portfolio_Return_nsei'].sum()
a3_in=case3_data_in_nsei['Portfolio_Return_nsei'].sum()
a4_in=case4_data_in_nsei['Portfolio_Return_nsei'].sum()
a5_in=case5_data_in_nsei['Portfolio_Return_nsei'].sum()
a_in=(a1_in+a2_in+a3_in+a4_in+a5_in)

b1_in=case1_data_in_nsei['Portfolio_Return_IN_bond'].sum()
b2_in=case2_data_in_nsei['Portfolio_Return_IN_bond'].sum()
b3_in=case3_data_in_nsei['Portfolio_Return_IN_bond'].sum()
b4_in=case4_data_in_nsei['Portfolio_Return_IN_bond'].sum()
b5_in=case5_data_in_nsei['Portfolio_Return_IN_bond'].sum()
b_in=(b1_in+b2_in+b3_in+b4_in+b5_in)

c1_in=filt_case1_nsei['^NSEI'].sum()
c2_in=filt_case2_nsei['^NSEI'].sum()
c3_in=filt_case3_nsei['^NSEI'].sum()
c4_in=filt_case4_nsei['^NSEI'].sum()
c5_in=filt_case5_nsei['^NSEI'].sum()
c_in=(c1_in+c2_in+c3_in+c4_in+c5_in)

d_in=return_multiData['^NSEI'].sum()
final_in=a_in+b_in+c_in-d_in
final_in_usd=final_in*1000000
final_in_usd_per_change=((final_in_usd-1000000)/1000000)*100
final_in_usd_per_change

701.2344161308868

FOR CHINA

filtered_dates1 = ann_vol_portfolio[((ann_vol_portfolio['^HSI'] < 15) & (ann_vol_portfolio['^HSI'].diff() <10))]['DateTime']
filtered_dates4  = ann_vol_portfolio[(((27 > ann_vol_portfolio['^HSI']) & (ann_vol_portfolio['^HSI'] >= 25)) | ((15 > ann_vol_portfolio['^HSI'].diff()) & (ann_vol_portfolio['^HSI'].diff() > 10)))]['DateTime']
filtered_dates3  = ann_vol_portfolio[(((25 > ann_vol_portfolio['^HSI']) & (ann_vol_portfolio['^HSI'] >= 20)) | ((30 > ann_vol_portfolio['^HSI'].diff()) & (ann_vol_portfolio['^HSI'].diff() > 20)))]['DateTime']
filtered_dates2  = ann_vol_portfolio[(((20 > ann_vol_portfolio['^HSI']) & (ann_vol_portfolio['^HSI'] >= 15)) | ((20 > ann_vol_portfolio['^HSI'].diff()) & (ann_vol_portfolio['^HSI'].diff() > 15)))]['DateTime']
filtered_dates5  = ann_vol_portfolio[((ann_vol_portfolio['^HSI']  >=  27) | (ann_vol_portfolio['^HSI'].diff() >25))]['DateTime']

# Dates fullfilling the condition
def selected_returns_dataframe(filtered_dates):
  selected_returns = []
  for date in filtered_dates:
    return_value = df_bond_return.loc[df_bond_return['Date'] == date, 'CHI_Price'].values
    if len(return_value) > 0:
        selected_returns.append((date, return_value[0]))

  selected_return=pd.DataFrame(selected_returns)
  selected_return.set_axis(['Date', 'Portfolio_Return_chi_bond'], axis=1, inplace=True)
  return selected_return

CASE 1

case1_chi=selected_returns_dataframe(filtered_dates1)
case1_chi.head()

	Date	Portfolio_Return_chi_bond
0	2015-03-20	0.006814
1	2015-03-23	0.001137
2	2015-03-24	0.002279
3	2015-03-25	-0.006229
4	2015-03-26	-0.008422

# Histogram of bond
case1_chi.plot.hist()

<Axes: ylabel='Frequency'>

selected_returns_bond_chi=case1_chi.set_index('Date')
selected_returns_bond_chi.plot()
plt.ylabel('Return')

Text(0, 0.5, 'Return')

case1_hsi=case1_hsi.set_index('Date')
case1_chi=case1_chi.set_index('Date')

case1_data_chi_hsi = case1_hsi.join(case1_chi, how='inner')
case1_data_chi_hsi.head()

	Portfolio_Return_hsi	Portfolio_Return_chi_bond
Date
2015-03-20	-0.003827	0.006814
2015-03-23	0.004893	0.001137
2015-03-24	-0.003875	0.002279
2015-03-25	0.005272	-0.006229
2015-03-26	-0.001270	-0.008422

# chi
weighted_return_c1_chi = (wts_c1 * case1_data_chi_hsi)

#Cumulative Portfolio chi
portfolio_return_c1_chi = weighted_return_c1_chi.sum(axis=1)
cumprod_portfolio_c1_chi = (1 + portfolio_return_c1_chi).cumprod()

#Plotting the solution portfolio's returns
plt.figure(figsize=(10, 5))
plt.plot(cumprod_portfolio_c1_chi, label='CASE 1(90E+10B)')
plt.legend(loc='upper left', fontsize=15)
plt.title("china Portfolio Performance C1")
plt.ylabel('Cumulative Return')

Text(0, 0.5, 'Cumulative Return')

CASE 2

case2_chi=selected_returns_dataframe(filtered_dates2)
case2_chi.head()

	Date	Portfolio_Return_chi_bond
0	2013-01-04	0.005571
1	2013-01-07	-0.002778
2	2013-01-08	0.005587
3	2013-01-09	-0.002786
4	2013-01-10	-0.002778

# Histogram of bond
case2_chi.plot.hist()

<Axes: ylabel='Frequency'>

selected_returns_bond_chi=case2_chi.set_index('Date')
selected_returns_bond_chi.plot()
plt.ylabel('Return')

Text(0, 0.5, 'Return')

case2_hsi=case2_hsi.set_index('Date')
case2_chi=case2_chi.set_index('Date')

case2_data_chi_hsi = case2_hsi.join(case2_chi, how='inner')
case2_data_chi_hsi.head()

	Portfolio_Return_hsi	Portfolio_Return_chi_bond
Date
2013-01-04	-0.002885	0.005571
2013-01-07	-0.000057	-0.002778
2013-01-08	-0.009368	0.005587
2013-01-09	0.004642	-0.002786
2013-01-10	0.005851	-0.002778

# china
weighted_return_c2_chi = (wts_c2 * case2_data_chi_hsi)

#Cumulative Portfolio chi
portfolio_return_c2_chi = weighted_return_c2_chi.sum(axis=1)
cumprod_portfolio_c2_chi = (1 + portfolio_return_c2_chi).cumprod()

#Plotting the solution portfolio's returns
plt.figure(figsize=(10, 5))
plt.plot(cumprod_portfolio_c2_chi, label='CASE 2(80E+20B)')
plt.legend(loc='upper left', fontsize=15)
plt.title("china Portfolio Performance C2")
plt.ylabel('Cumulative Return')

Text(0, 0.5, 'Cumulative Return')

CASE 3

case3_chi=selected_returns_dataframe(filtered_dates3)
case3_chi.head()

	Date	Portfolio_Return_chi_bond
0	2015-09-09	0.003851
1	2015-09-10	0.000889
2	2015-09-11	0.003570
3	2015-09-14	0.000595
4	2015-09-15	0.005388

# Histogram of bond
case3_chi.plot.hist()

<Axes: ylabel='Frequency'>

selected_returns_bond_chi=case3_chi.set_index('Date')
selected_returns_bond_chi.plot()
plt.ylabel('Return')

Text(0, 0.5, 'Return')

case3_hsi=case3_hsi.set_index('Date')
case3_chi=case3_chi.set_index('Date')

case3_data_chi_hsi = case3_hsi.join(case3_chi, how='inner')
case3_data_chi_hsi.head()

	Portfolio_Return_hsi	Portfolio_Return_chi_bond
Date
2015-09-09	0.041031	0.003851
2015-09-10	-0.025702	0.000889
2015-09-11	-0.002696	0.003570
2015-09-14	0.002675	0.000595
2015-09-15	-0.004947	0.005388

# china
weighted_return_c3_chi = (wts_c3 * case3_data_chi_hsi)

#Cumulative Portfolio chi

portfolio_return_c3_chi = weighted_return_c3_chi.sum(axis=1)
cumprod_portfolio_c3_chi = (1 + portfolio_return_c3_chi).cumprod()

#Plotting the solution portfolio's returns
plt.figure(figsize=(10, 5))
plt.plot(cumprod_portfolio_c3_chi, label='CASE 3(70E+30B)')
plt.legend(loc='upper left', fontsize=15)
plt.title("china Portfolio Performance C3")
plt.ylabel('Cumulative Return')

Text(0, 0.5, 'Cumulative Return')

CASE 4

case4_chi=selected_returns_dataframe(filtered_dates4)
case4_chi.head()

	Date	Portfolio_Return_chi_bond
0	2020-08-20	-0.000666
1	2020-08-21	0.000666
2	2020-08-24	-0.015410
3	2020-08-25	-0.002942
4	2020-08-26	-0.002608

# Histogram of bond
case4_chi.plot.hist()

<Axes: ylabel='Frequency'>

selected_returns_bond_chi=case4_chi.set_index('Date')
selected_returns_bond_chi.plot()
plt.ylabel('Return')

Text(0, 0.5, 'Return')

case4_hsi=case4_hsi.set_index('Date')
case4_chi=case4_chi.set_index('Date')

case4_data_chi_hsi = case4_hsi.join(case4_chi, how='inner')
case4_data_chi_hsi.head()

	Portfolio_Return_hsi	Portfolio_Return_chi_bond
Date
2020-08-20	-0.015391	-0.000666
2020-08-21	0.013006	0.000666
2020-08-24	0.017430	-0.015410
2020-08-25	-0.002558	-0.002942
2020-08-26	0.000218	-0.002608

# china
weighted_return_c4_chi = (wts_c4 * case4_data_chi_hsi)

#Cumulative Portfolio chi
portfolio_return_c4_chi = weighted_return_c4_chi.sum(axis=1)
cumprod_portfolio_c4_chi = (1 + portfolio_return_c4_chi).cumprod()

#Plotting the solution portfolio's returns
plt.figure(figsize=(10, 5))
plt.plot(cumprod_portfolio_c4_chi, label='CASE 4(60E+40B)')
plt.legend(loc='upper left', fontsize=15)
plt.title("china Portfolio Performance C4")
plt.ylabel('Cumulative Return')

Text(0, 0.5, 'Cumulative Return')

CASE 5

case5_chi=selected_returns_dataframe(filtered_dates5)
case5_chi.head()

	Date	Portfolio_Return_chi_bond
0	2022-05-19	-0.003898
1	2022-05-20	0.006061
2	2022-05-23	0.001428
3	2022-05-24	0.006830
4	2022-05-25	0.012741

# Histogram of bond
case5_chi.plot.hist()

<Axes: ylabel='Frequency'>

selected_returns_bond_chi=case5_chi.set_index('Date')
selected_returns_bond_chi.plot()
plt.ylabel('Return')

Text(0, 0.5, 'Return')

case5_hsi=case5_hsi.set_index('Date')
case5_chi=case5_chi.set_index('Date')

case5_data_chi_hsi = case5_hsi.join(case5_chi, how='inner')
case5_data_chi_hsi.head()

	Portfolio_Return_hsi	Portfolio_Return_chi_bond
Date
2022-05-19	-0.025363	-0.003898
2022-05-20	0.029649	0.006061
2022-05-23	-0.011931	0.001428
2022-05-24	-0.017487	0.006830
2022-05-25	0.002942	0.012741

# china
weighted_return_c5_chi = (wts_c5 * case5_data_chi_hsi)

#Cumulative Portfolio china
portfolio_return_c5_chi = weighted_return_c5_chi.sum(axis=1)
cumprod_portfolio_c5_chi = (1 + portfolio_return_c5_chi).cumprod()

#Plotting the solution portfolio's returns
plt.figure(figsize=(10, 5))
plt.plot(cumprod_portfolio_c5_chi, label='CASE 5(10E+90B)')
plt.legend(loc='upper left', fontsize=15)
plt.title("china Portfolio Performance C5")
plt.ylabel('Cumulative Return')

Text(0, 0.5, 'Cumulative Return')

EQUITY

drawdown(case1_data_chi_hsi['Portfolio_Return_hsi'])

	Wealth	Previous Peak	Drawdown
Date
2015-03-20	996.172675	996.172675	0.000000
2015-03-23	1001.047009	1001.047009	0.000000
2015-03-24	997.168200	1001.047009	-0.003875
2015-03-25	1002.425114	1002.425114	0.000000
2015-03-26	1001.152053	1002.425114	-0.001270
...	...	...	...
2018-02-28	1208.089624	1329.979345	-0.091648
2018-03-01	1215.904552	1329.979345	-0.085772
2018-03-05	1170.554883	1329.979345	-0.119870
2018-03-06	1195.008289	1329.979345	-0.101484
2018-03-07	1182.717328	1329.979345	-0.110725

182 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case2_data_chi_hsi['Portfolio_Return_hsi'])

	Wealth	Previous Peak	Drawdown
Date
2013-01-04	997.114795	997.114795	0.000000
2013-01-07	997.057533	997.114795	-0.000057
2013-01-08	987.716780	997.114795	-0.009425
2013-01-09	992.301723	997.114795	-0.004827
2013-01-10	998.107192	998.107192	0.000000
...	...	...	...
2021-08-17	589.486037	1137.842493	-0.481927
2021-08-18	592.259713	1137.842493	-0.479489
2021-08-26	585.845672	1137.842493	-0.485126
2021-08-27	585.665905	1137.842493	-0.485284
2021-09-08	584.939175	1137.842493	-0.485923

1016 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case3_data_chi_hsi['Portfolio_Return_hsi'])

	Wealth	Previous Peak	Drawdown
Date
2015-09-09	1041.030617	1041.030617	0.000000
2015-09-10	1014.274443	1041.030617	-0.025702
2015-09-11	1011.540036	1041.030617	-0.028328
2015-09-14	1014.246238	1041.030617	-0.025729
2015-09-15	1009.228611	1041.030617	-0.030549
...	...	...	...
2022-03-10	1270.194730	1671.907147	-0.240272
2022-03-11	1249.736233	1671.907147	-0.252509
2022-03-14	1187.587520	1671.907147	-0.289681
2022-03-15	1119.695874	1671.907147	-0.330288
2022-03-16	1221.384364	1671.907147	-0.269466

709 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case4_data_chi_hsi['Portfolio_Return_hsi'])

	Wealth	Previous Peak	Drawdown
Date
2020-08-20	984.609360	984.609360	0.000000
2020-08-21	997.415682	997.415682	0.000000
2020-08-24	1014.800876	1014.800876	0.000000
2020-08-25	1012.205077	1014.800876	-0.002558
2020-08-26	1012.426229	1014.800876	-0.002340
...	...	...	...
2022-05-11	870.521733	1014.800876	-0.142175
2022-05-12	851.015016	1014.800876	-0.161397
2022-05-13	873.779913	1014.800876	-0.138964
2022-05-17	902.349744	1014.800876	-0.110811
2022-05-18	904.178739	1014.800876	-0.109009

86 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case5_data_chi_hsi['Portfolio_Return_hsi'])

	Wealth	Previous Peak	Drawdown
Date
2022-05-19	974.637060	974.637060	0.000000
2022-05-20	1003.534196	1003.534196	0.000000
2022-05-23	991.560919	1003.534196	-0.011931
2022-05-24	974.221445	1003.534196	-0.029210
2022-05-25	977.087611	1003.534196	-0.026353
...	...	...	...
2022-12-29	928.392618	1085.965288	-0.145099
2022-12-30	930.286427	1085.965288	-0.143355
2023-01-03	947.399039	1085.965288	-0.127597
2023-01-04	977.864938	1085.965288	-0.099543
2023-01-05	990.048120	1085.965288	-0.088324

138 rows × 3 columns

Values

Distributions

2-d distributions

Time series

BOND

drawdown(case1_data_chi_hsi['Portfolio_Return_chi_bond'])

	Wealth	Previous Peak	Drawdown
Date
2015-03-20	1006.814310	1006.814310	0.000000
2015-03-23	1007.959068	1007.959068	0.000000
2015-03-24	1010.256410	1010.256410	0.000000
2015-03-25	1003.963760	1010.256410	-0.006229
2015-03-26	995.508141	1010.256410	-0.014599
...	...	...	...
2018-02-28	929.687692	1028.542292	-0.096111
2018-03-01	927.281045	1028.542292	-0.098451
2018-03-05	924.170970	1028.542292	-0.101475
2018-03-06	926.801210	1028.542292	-0.098918
2018-03-07	930.170521	1028.542292	-0.095642

182 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case2_data_chi_hsi['Portfolio_Return_chi_bond'])

	Wealth	Previous Peak	Drawdown
Date
2013-01-04	1005.571031	1005.571031	0.000000
2013-01-07	1002.777778	1005.571031	-0.002778
2013-01-08	1008.379888	1008.379888	0.000000
2013-01-09	1005.571031	1008.379888	-0.002786
2013-01-10	1002.777778	1008.379888	-0.005556
...	...	...	...
2021-08-17	1254.791379	1274.636580	-0.015569
2021-08-18	1253.471935	1274.636580	-0.016604
2021-08-26	1241.837785	1274.636580	-0.025732
2021-08-27	1262.672533	1274.636580	-0.009386
2021-09-08	1260.916382	1274.636580	-0.010764

1016 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case3_data_chi_hsi['Portfolio_Return_chi_bond'])

	Wealth	Previous Peak	Drawdown
Date
2015-09-09	1003.850711	1003.850711	0.000000
2015-09-10	1004.743552	1004.743552	0.000000
2015-09-11	1008.330854	1008.330854	0.000000
2015-09-14	1008.931230	1008.931230	0.000000
2015-09-15	1014.366956	1014.366956	0.000000
...	...	...	...
2022-03-10	1164.155310	1347.260025	-0.135909
2022-03-11	1182.162029	1347.260025	-0.122544
2022-03-14	1163.743074	1347.260025	-0.136215
2022-03-15	1169.541082	1347.260025	-0.131911
2022-03-16	1168.293793	1347.260025	-0.132837

709 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case4_data_chi_hsi['Portfolio_Return_chi_bond'])

	Wealth	Previous Peak	Drawdown
Date
2020-08-20	999.334443	999.334443	0.000000
2020-08-21	1000.000000	1000.000000	0.000000
2020-08-24	984.590164	1000.000000	-0.015410
2020-08-25	981.693364	1000.000000	-0.018307
2020-08-26	979.132703	1000.000000	-0.020867
...	...	...	...
2022-05-11	968.394084	1000.000000	-0.031606
2022-05-12	967.363147	1000.000000	-0.032637
2022-05-13	965.991973	1000.000000	-0.034008
2022-05-17	970.808172	1000.000000	-0.029192
2022-05-18	969.772091	1000.000000	-0.030228

86 rows × 3 columns

Values

Distributions

2-d distributions

Time series

drawdown(case5_data_chi_hsi['Portfolio_Return_chi_bond'])

	Wealth	Previous Peak	Drawdown
Date
2022-05-19	996.102055	996.102055	0.000000
2022-05-20	1002.139037	1002.139037	0.000000
2022-05-23	1003.570154	1003.570154	0.000000
2022-05-24	1010.424155	1010.424155	0.000000
2022-05-25	1023.298143	1023.298143	0.000000
...	...	...	...
2022-12-29	1011.508183	1104.248944	-0.083985
2022-12-30	1015.380995	1104.248944	-0.080478
2023-01-03	1018.927558	1104.248944	-0.077266
2023-01-04	1016.442369	1104.248944	-0.079517
2023-01-05	1016.442369	1104.248944	-0.079517

138 rows × 3 columns

Values

Distributions

2-d distributions

Time series

a1_chi=case1_data_chi_hsi['Portfolio_Return_hsi'].sum()
a2_chi=case2_data_chi_hsi['Portfolio_Return_hsi'].sum()
a3_chi=case3_data_chi_hsi['Portfolio_Return_hsi'].sum()
a4_chi=case4_data_chi_hsi['Portfolio_Return_hsi'].sum()
a5_chi=case5_data_chi_hsi['Portfolio_Return_hsi'].sum()
a_chi=(a1_chi+a2_chi+a3_chi+a4_chi+a5_chi)

b1_chi=case1_data_chi_hsi['Portfolio_Return_chi_bond'].sum()
b2_chi=case2_data_chi_hsi['Portfolio_Return_chi_bond'].sum()
b3_chi=case3_data_chi_hsi['Portfolio_Return_chi_bond'].sum()
b4_chi=case4_data_chi_hsi['Portfolio_Return_chi_bond'].sum()
b5_chi=case5_data_chi_hsi['Portfolio_Return_chi_bond'].sum()
b_chi=(b1_chi+b2_chi+b3_chi+b4_chi+b5_chi)

c1_chi=filt_case1_hsi['^HSI'].sum()
c2_chi=filt_case2_hsi['^HSI'].sum()
c3_chi=filt_case3_hsi['^HSI'].sum()
c4_chi=filt_case4_hsi['^HSI'].sum()
c5_chi=filt_case5_hsi['^HSI'].sum()
c_chi=(c1_chi+c2_chi+c3_chi+c4_chi+c5_chi)

d_chi=return_multiData['^HSI'].sum()
final_chi=a_chi+b_chi+c_chi-d_chi
final_chi_usd=final_chi*1000000
final_chi_usd_per_change=((final_chi_usd-1000000)/1000000)*100
final_chi_usd_per_change

144.09711313419874

Returns of an investments of $1000000 from different markets

import matplotlib.pyplot as plt
import numpy as np
data = {
    'Indices': ['^FTSE', '^GSPC', '^HSI', '^NSEI'],
    'USD_Value': [final_uk_usd, final_us_usd, final_chi_usd, final_in_usd],
    'Percentage_change':[final_uk_usd_per_change,final_us_usd_per_change,final_chi_usd_per_change,final_in_usd_per_change]
}

df=pd.DataFrame(data)
# Calculate the width of each bar
bar_width = 0.35

# Create a figure and primary y-axis (left)
fig, ax1 = plt.subplots()

# Create an array of x positions for the bars
x = np.arange(len(df['Indices']))

# Plot the first bar graph on the primary y-axis
ax1.bar(x - bar_width/2, df['USD_Value'], bar_width, color='orange', label='Return (usd in million)')
ax1.set_xlabel('Indices')
ax1.set_ylabel('USD in million', color='orange')
ax1.tick_params(axis='y', labelcolor='orange')

# Create a secondary y-axis (right)
ax2 = ax1.twinx()

# Plot the second bar graph on the secondary y-axis
ax2.bar(x + bar_width/2, df['Percentage_change'], bar_width, color='cyan', label='Percentage Change', alpha=0.7)
ax2.set_ylabel('% Change', color='cyan')
ax2.tick_params(axis='y', labelcolor='cyan')

# Adjust x-axis positions and labels
ax1.set_xticks(x)
ax1.set_xticklabels(df['Indices'])

# Combine legends from both axes
lines1, labels1 = ax1.get_legend_handles_labels()
lines2, labels2 = ax2.get_legend_handles_labels()
lines = lines1 + lines2
labels = labels1 + labels2

# Display a single legend for both bar graphs
ax1.legend(lines, labels, loc='upper right')

plt.title('Return and % Change')
plt.tight_layout()
plt.show()

Cumulative returns during Pandemic year of 2020

Equity

Tickers_P = [ '^FTSE', '^GSPC', '^HSI','^NSEI','GLD']
Start_Date1 = "2020-01-01"
End_Date1 = "2020-12-31"

CovidData = GetAdjCloseQuotes(Tickers_P, Start_Date1, End_Date1)
CovidData.head()

	^FTSE	^GSPC	^HSI	^NSEI	GLD
DateTime
2020-01-02	7604.299805	3257.850098	28543.519531	12282.200195	143.949997
2020-01-03	7622.399902	3234.850098	28451.500000	12226.650391	145.860001
2020-01-06	7575.299805	3246.280029	28226.189453	11993.049805	147.389999
2020-01-07	7573.899902	3237.179932	28322.060547	12052.950195	147.970001
2020-01-08	7574.899902	3253.050049	28087.919922	12025.349609	146.860001

# Calculating daily returns
return_CovidData = CovidData.pct_change().dropna()
return_CovidData

	^FTSE	^GSPC	^HSI	^NSEI	GLD
DateTime
2020-01-03	0.002380	-0.007060	-0.003224	-0.004523	0.013269
2020-01-06	-0.006179	0.003533	-0.007919	-0.019106	0.010489
2020-01-07	-0.000185	-0.002803	0.003397	0.004995	0.003935
2020-01-08	0.000132	0.004902	-0.008267	-0.002290	-0.007502
2020-01-09	0.003063	0.006655	0.016843	0.015846	-0.005652
...	...	...	...	...	...
2020-12-22	0.000000	-0.002073	-0.007125	0.010346	-0.007903
2020-12-23	0.012390	0.000746	0.008570	0.010010	0.006648
2020-12-24	0.000970	0.003537	0.001650	0.010893	0.003985
2020-12-29	0.015472	0.006476	0.006895	0.013335	0.000000
2020-12-30	-0.007103	0.001342	0.021778	0.003542	0.007655

230 rows × 5 columns

# Cumulative return
cumprod_return_CovidData = (1 + return_CovidData).cumprod()
cumprod_return_CovidData.head()

	^FTSE	^GSPC	^HSI	^NSEI	GLD
DateTime
2020-01-03	1.002380	0.992940	0.996776	0.995477	1.013269
2020-01-06	0.996186	0.996449	0.988883	0.976458	1.023897
2020-01-07	0.996002	0.993655	0.992241	0.981335	1.027926
2020-01-08	0.996134	0.998527	0.984038	0.979088	1.020215
2020-01-09	0.999185	1.005172	1.000612	0.994602	1.014449

fig, axes = plt.subplots(1, 2, sharex=True, figsize=(20,5))

sns.lineplot(data=return_CovidData, dashes=False, ax=axes[0])
axes[0].set_ylabel('Return')
axes[0].set_title('Return of Equity During Covid')


sns.lineplot(data=cumprod_return_CovidData, dashes=False, ax=axes[1])
axes[1].set_ylabel('Cumulative return')
axes[1].set_title('Cumulative Returns of Equity During Covid')

Text(0.5, 1.0, 'Cumulative Returns of Equity During Covid')

Bond

bond_CovidData=df_bond.loc[Start_Date1:End_Date1]
bond_CovidData.head()

	US_Price	IN_Price	UK_Price	CHI_Price
Date
2020-12-31	0.916	5.894	0.196	3.203
2020-12-30	0.926	5.899	0.211	3.192
2020-12-29	0.940	5.927	0.213	3.205
2020-12-28	0.923	5.925	0.253	3.213
2020-12-24	0.926	5.929	0.257	3.259

# Return
bond_return_CovidData=bond_CovidData.pct_change().dropna()
bond_return_CovidData.head()

	US_Price	IN_Price	UK_Price	CHI_Price
Date
2020-12-30	0.010917	0.000848	0.076531	-0.003434
2020-12-29	0.015119	0.004747	0.009479	0.004073
2020-12-28	-0.018085	-0.000337	0.187793	0.002496
2020-12-24	0.003250	0.000675	0.015810	0.014317
2020-12-23	0.020518	0.003542	0.108949	-0.004603

# Cumulative Return
cumprod_cum_return_CovidData=(1+bond_return_CovidData).cumprod()
cumprod_cum_return_CovidData.head()

	US_Price	IN_Price	UK_Price	CHI_Price
Date
2020-12-30	1.010917	1.000848	1.076531	0.996566
2020-12-29	1.026201	1.005599	1.086735	1.000624
2020-12-28	1.007642	1.005260	1.290816	1.003122
2020-12-24	1.010917	1.005938	1.311224	1.017484
2020-12-23	1.031659	1.009501	1.454082	1.012800

fig, axes = plt.subplots(1, 2, sharex=True, figsize=(20,5))

sns.lineplot(data=bond_return_CovidData, dashes=False, ax=axes[0])
axes[0].set_ylabel('Return')
axes[0].set_title('Return of Bond During Covid')


sns.lineplot(data=cumprod_cum_return_CovidData, dashes=False, ax=axes[1])
axes[1].set_ylabel('Cumulative return')
axes[1].set_title('Cumulative Returns of Bond During Covid')

Text(0.5, 1.0, 'Cumulative Returns of Bond During Covid')